chemistry_predict_logP_tensorflow.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "604baa65",
   "metadata": {},
   "source": [
    "# Deep Learning for Chemistry\n",
    "\n",
    "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/tonyreina/chemistry/blob/main/chemistry_predict_logP_tensorflow.ipynb)\n",
    "\n",
    "This notebook is based on the excellent [Kaggle tutorial](https://www.kaggle.com/vladislavkisin/tutorial-ml-in-chemistry-research-rdkit-mol2vec) from [Vlad Kisin](https://www.kaggle.com/vladislavkisin). In this example, you'll learn how to read a Chemistry datafile and create predictive models of [lipophilicity](https://en.wikipedia.org/wiki/Partition_coefficient#Partition_coefficient_and_log_P). \n",
    "\n",
    "Lipophilicity is the ability of a chemical compound to dissolve in non-polar (fatty or oily) solvents. In simple terms, if you had a glass of oil and water (which will separate with one on top of the other as in the figure below), then lipophilicity is the proportion of how much a chemical dissovles in the water portion versus the oil portion. In the figure below there are 3 molecules in water to every 1 molecule in oil. P is 3 and the log P is $\\log_{10}{3} = 0.477$.\n",
    "\n",
    "Lipophilicity contributes to the absorption, distribution, metabolism, excretion, and toxicity of a pharmaceutical and contributes to a drug's potency and selectivity.\n",
    "\n",
    "![Figure1](logP.png)\n",
    "\n",
    "I'll demonstrate how to load the raw data from a CSV file and use the [RD-Kit](https://github.com/rdkit/rdkit) and [Mol2Vec](https://pubs.acs.org/doi/10.1021/acs.jcim.7b00616) packages to create features based on the chemical formula of a molecule."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e90dfac",
   "metadata": {},
   "source": [
    "## Import the usual libraries\n",
    "\n",
    "You might need to install [Seaborn](https://seaborn.pydata.org/) via `pip install seaborn` or `conda install seaborn` and [Pandas](https://pandas.pydata.org/) via `pip install pandas` or `conda install pandas`\n",
    "\n",
    "Here are the commands I used to create the conda environment:\n",
    "\n",
    "```\n",
    "conda create -n chem python=3.8 pip jupyter matplotlib seaborn\n",
    "conda activate chem\n",
    "conda install -c conda-forge rdkit\n",
    "pip install git+https://github.com/samoturk/mol2vec\n",
    "wget https://raw.githubusercontent.com/tonyreina/mol2vec/master/mol2vec/features.py -O  ~/anaconda3/envs/chem/lib/python3.8/site-packages/mol2vec/features.py\n",
    "wget https://github.com/samoturk/mol2vec_notebooks/blob/master/Notebooks/model_300dim.pkl\n",
    "pip install -U tensorflow==2.4.1\n",
    "pip install openvino-tensorflow==0.5.0\n",
    "conda install scikit-learn\n",
    "pip install py3Dmol\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e056e1f0",
   "metadata": {},
   "source": [
    "### Next cell needed for running in Google Colab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5f52b354",
   "metadata": {},
   "outputs": [],
   "source": [
    "%pip install rdkit-pypi\n",
    "%pip install py3Dmol\n",
    "%pip install git+https://github.com/samoturk/mol2vec\n",
    "%pip install openvino_tensorflow\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7bdffb0d",
   "metadata": {},
   "source": [
    "## Important!\n",
    "Currently (MAY 2021) the [OpenVINO™-TensorFlow bridge](https://github.com/openvinotoolkit/openvino_tensorflow) only works with Python 3.8 and TensorFlow 2.4.1. You may want to check the [website](https://github.com/openvinotoolkit/openvino_tensorflow) to see what version of TensorFlow and Python you need to use. \n",
    "\n",
    "You only need the OpenVINO™-TensorFlow bridge and the specific version of TensorFlow if you want to complete the last section of this tutorial. Otherwise, any current version of TensorFlow should work."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "15db48b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import numpy as np                #!pip install numpy\n",
    "import pandas as pd               #!pip install pandas\n",
    "import matplotlib.pyplot as plt   #!pip install matplotlib\n",
    "import seaborn as sns             #!pip install seaborn"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "2891707a",
   "metadata": {},
   "source": [
    "## Load the raw data from the CSV file for the lipophilicity dataset\n",
    "\n",
    "The lipophilicity dataset is available on [Kaggle](https://www.kaggle.com/matthewmasters/chemical-structure-and-logp) and released under the Public Domain ([CC0](https://creativecommons.org/share-your-work/public-domain/cc0/)). The raw data is in a CSV file with the SMILES notation of the chemical in the first column and the lipophilicity (logP) in the second column.\n",
    "\n",
    "[SMILES](https://en.wikipedia.org/wiki/Simplified_molecular-input_line-entry_system) stands for the **S**implified **M**olecular **I**nput **L**ine **E**ntry **S**ystem which is a method for describing the structure of chemical species using short ASCII strings. Essentially it's a short-hand notation for describing a molecule including information about which atoms are connected together, the types of chemical bonds, and the 2D structure. In SMILES notation, 'C' represents a carbon atom, 'N' represents a nitrogen atom, 'Cl' represents a Chlorine atom, and '=' indicates a double bond between the adjacent atoms. For example, in the figure below the SMILES notation for the top/left molecule is **Cl\n",
    "ClC\\[C@H\\](\\[C@@H\\](C)Cl)Cl**.\n",
    "\n",
    "A good resource for learning about SMILES notation is:\n",
    "\n",
    "Quirós, M., Gražulis, S., Girdzijauskaitė, S. et al. Using SMILES strings for the description of chemical connectivity in the Crystallography Open Database. J Cheminform 10, 23 (2018). https://doi.org/10.1186/s13321-018-0279-6\n",
    "\n",
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a85d557b",
   "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>smiles</th>\n",
       "      <th>logP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C[C@H]([C@@H](C)Cl)Cl</td>\n",
       "      <td>2.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C(C=CBr)N</td>\n",
       "      <td>0.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CCC(CO)Br</td>\n",
       "      <td>1.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[13CH3][13CH2][13CH2][13CH2][13CH2][13CH2]O</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CCCOCCP</td>\n",
       "      <td>0.6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        smiles  logP\n",
       "0                        C[C@H]([C@@H](C)Cl)Cl   2.3\n",
       "1                                    C(C=CBr)N   0.3\n",
       "2                                    CCC(CO)Br   1.3\n",
       "3  [13CH3][13CH2][13CH2][13CH2][13CH2][13CH2]O   2.0\n",
       "4                                      CCCOCCP   0.6"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('https://raw.githubusercontent.com/tonyreina/chemistry/main/logP_dataset.csv', names=['smiles', 'logP']) #load the CSV file\n",
    "df.head()  # Display a few rows"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38d414e7",
   "metadata": {},
   "source": [
    "## What is our goal?\n",
    "\n",
    "We'd like to create an algorithm **f(x)** that takes the **smiles** data as input and returns the **logP** as the output. For example,\n",
    "\n",
    "```\n",
    "f( 'C[C@H]([C@@H](C)Cl)Cl' ) = 2.3\n",
    "f( 'CCCOCCP' )               = 0.6\n",
    "```\n",
    "\n",
    "We can accomplish this with machine learning."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7eae45a",
   "metadata": {},
   "source": [
    "## RD-Kit\n",
    "[RD-Kit](https://github.com/rdkit/rdkit) is an open-source (BSD-3) Python package for data scientists to work with Chemistry data. It has some excellent modules that allow us to preprocess data from chemistry experiments.\n",
    "\n",
    "You can install it using the conda package:\n",
    "`conda install -c rdkit rdkit`\n",
    "\n",
    "With RD-Kit we can transform the SMILES notation into a more useful Python object called a *Mol* (for molecule). Mol objects contain the atoms, bonds, connectivity and coordinates of a molecule. It adds important 3D information, such as the distance between atoms, that may be useful to our model. We can use *Mol* objects to craft features of the chemical to provide as inputs to our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ff308042",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "RDKit WARNING: [10:44:17] Enabling RDKit 2019.09.3 jupyter extensions\n"
     ]
    },
    {
     "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>smiles</th>\n",
       "      <th>logP</th>\n",
       "      <th>mol</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C[C@H]([C@@H](C)Cl)Cl</td>\n",
       "      <td>2.3</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc54158e3a0&gt;</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C(C=CBr)N</td>\n",
       "      <td>0.3</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc549ea3d00&gt;</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CCC(CO)Br</td>\n",
       "      <td>1.3</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc5411b5580&gt;</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[13CH3][13CH2][13CH2][13CH2][13CH2][13CH2]O</td>\n",
       "      <td>2.0</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc5411b5520&gt;</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CCCOCCP</td>\n",
       "      <td>0.6</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc5411b55e0&gt;</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        smiles  logP  \\\n",
       "0                        C[C@H]([C@@H](C)Cl)Cl   2.3   \n",
       "1                                    C(C=CBr)N   0.3   \n",
       "2                                    CCC(CO)Br   1.3   \n",
       "3  [13CH3][13CH2][13CH2][13CH2][13CH2][13CH2]O   2.0   \n",
       "4                                      CCCOCCP   0.6   \n",
       "\n",
       "                                                mol  \n",
       "0  <rdkit.Chem.rdchem.Mol object at 0x7fc54158e3a0>  \n",
       "1  <rdkit.Chem.rdchem.Mol object at 0x7fc549ea3d00>  \n",
       "2  <rdkit.Chem.rdchem.Mol object at 0x7fc5411b5580>  \n",
       "3  <rdkit.Chem.rdchem.Mol object at 0x7fc5411b5520>  \n",
       "4  <rdkit.Chem.rdchem.Mol object at 0x7fc5411b55e0>  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Importing Chem module\n",
    "from rdkit import Chem \n",
    "\n",
    "#Method transforms smiles strings to mol rdkit object\n",
    "df['mol'] = df['smiles'].apply(lambda x: Chem.MolFromSmiles(x)) \n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85a61a48",
   "metadata": {},
   "source": [
    "## Plot some chemical structures\n",
    "\n",
    "Once we have our molecules as *Mol* Python objects, then RD-Kit allows us to plot the chemical structures easily."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b6080e2f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": "<svg xmlns=\"http://www.w3.org/2000/svg\" xmlns:rdkit=\"http://www.rdkit.org/xml\" xmlns:xlink=\"http://www.w3.org/1999/xlink\" version=\"1.1\" baseProfile=\"full\" xml:space=\"preserve\" width=\"1000px\" height=\"800px\" viewBox=\"0 0 1000 800\">\n<!-- END OF HEADER -->\n<rect style=\"opacity:1.0;fill:#FFFFFF;stroke:none\" width=\"1000\" height=\"800\" x=\"0\" y=\"0\"> </rect>\n<path class=\"bond-0\" d=\"M 114.967,100 L 132.527,75.5723 L 127.343,72.5788 Z\" style=\"fill:#000000;fill-rule:evenodd;fill-opacity=1;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;\"/>\n<path class=\"bond-1\" d=\"M 114.967,100 L 85.0325,100\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 114.967,100 L 121.011,110.468\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 121.011,110.468 L 127.054,120.935\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 85.0325,100 L 67.4726,124.428 L 72.6575,127.421 Z\" style=\"fill:#000000;fill-rule:evenodd;fill-opacity=1;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;\"/>\n<path class=\"bond-3\" d=\"M 85.0325,100 L 78.989,89.5324\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 78.989,89.5324 L 72.9455,79.0647\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"64.411\" y=\"79.0647\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#00CC00\"><tspan>Cl</tspan></text>\n<text x=\"124.281\" y=\"130.914\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#00CC00\"><tspan>Cl</tspan></text>\n<text x=\"15.6615\" y=\"194\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C[C@H]([C@@H](C)Cl)Cl</tspan></text>\n<path class=\"bond-0\" d=\"M 274.076,108.98 L 300,94.013\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 274.076,108.98 L 265.866,104.241\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 265.866,104.241 L 257.657,99.5011\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 300,94.013 L 325.924,108.98\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 300.895,101.443 L 319.042,111.92\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 325.924,108.98 L 335.892,103.226\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 335.892,103.226 L 345.86,97.471\" style=\"fill:none;fill-rule:evenodd;stroke:#7F4C19;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"345.86\" y=\"99.0022\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#7F4C19\"><tspan>Br</tspan></text>\n<text x=\"237.837\" y=\"99.5011\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#0000FF\"><tspan>H</tspan><tspan style=\"baseline-shift:sub;font-size:6.75px;\">2</tspan><tspan>N</tspan></text>\n<text x=\"262.992\" y=\"194\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C(C=CBr)N</tspan></text>\n<path class=\"bond-0\" d=\"M 448.151,100 L 474.076,85.0325\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 474.076,85.0325 L 500,100\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 500,100 L 525.924,85.0325\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 500,100 L 500,112.473\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 500,112.473 L 500,124.946\" style=\"fill:none;fill-rule:evenodd;stroke:#7F4C19;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 525.924,85.0325 L 534.566,90.0217\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 534.566,90.0217 L 543.207,95.0108\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"542.865\" y=\"104.989\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>OH</tspan></text>\n<text x=\"494.011\" y=\"134.924\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#7F4C19\"><tspan>Br</tspan></text>\n<text x=\"461.594\" y=\"194\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>CCC(CO)Br</tspan></text>\n<path class=\"bond-0\" d=\"M 629.881,101.995 L 640.497,95.8664\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 655.806,95.8664 L 666.421,101.995\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 681.73,101.995 L 692.346,95.8664\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 707.654,95.8664 L 718.27,101.995\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 733.579,101.995 L 744.194,95.8664\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 759.503,95.8664 L 764.318,98.646\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 764.318,98.646 L 769.132,101.425\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"614.572\" y=\"111.903\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">13</tspan><tspan>C</tspan></text>\n<text x=\"640.497\" y=\"96.9352\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">13</tspan><tspan>C</tspan></text>\n<text x=\"666.421\" y=\"111.903\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">13</tspan><tspan>C</tspan></text>\n<text x=\"692.346\" y=\"96.9352\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">13</tspan><tspan>C</tspan></text>\n<text x=\"718.27\" y=\"111.903\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">13</tspan><tspan>C</tspan></text>\n<text x=\"744.194\" y=\"96.9352\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">13</tspan><tspan>C</tspan></text>\n<text x=\"768.789\" y=\"111.404\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>OH</tspan></text>\n<text x=\"535.875\" y=\"194\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>[13CH3][13CH2][13CH2][13CH2][13CH2][13CH2]O</tspan></text>\n<path class=\"bond-0\" d=\"M 822.227,106.415 L 848.151,91.4471\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 848.151,91.4471 L 874.076,106.415\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 874.076,106.415 L 884.708,100.276\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 884.708,100.276 L 895.34,94.1375\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 904.66,94.1375 L 915.292,100.276\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 915.292,100.276 L 925.924,106.415\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 925.924,106.415 L 951.849,91.4471\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 951.849,91.4471 L 960.058,96.1869\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 960.058,96.1869 L 968.268,100.927\" style=\"fill:none;fill-rule:evenodd;stroke:#FF7F00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"895.34\" y=\"96.4363\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>O</tspan></text>\n<text x=\"967.789\" y=\"111.903\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF7F00\"><tspan>PH</tspan><tspan style=\"baseline-shift:sub;font-size:6.75px;\">2</tspan><tspan/></text>\n<text x=\"863.589\" y=\"194\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>CCCOCCP</tspan></text>\n<path class=\"bond-0\" d=\"M 82.717,314.967 L 108.641,300\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 82.717,314.967 L 71.5846,308.54\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 71.5846,308.54 L 60.4521,302.113\" style=\"fill:none;fill-rule:evenodd;stroke:#33CCCC;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 108.641,300 L 119.774,293.573\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 119.774,293.573 L 130.906,287.145\" style=\"fill:none;fill-rule:evenodd;stroke:#33CCCC;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 108.641,300 L 114.685,310.468\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 114.685,310.468 L 120.728,320.935\" style=\"fill:none;fill-rule:evenodd;stroke:#33CCCC;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 108.641,300 L 102.598,289.532\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 102.598,289.532 L 96.5545,279.065\" style=\"fill:none;fill-rule:evenodd;stroke:#33CCCC;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"130.906\" y=\"290.022\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#33CCCC\"><tspan>F</tspan></text>\n<text x=\"119.949\" y=\"330.914\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#33CCCC\"><tspan>F</tspan></text>\n<text x=\"90.0145\" y=\"279.065\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#33CCCC\"><tspan>F</tspan></text>\n<text x=\"53.1331\" y=\"304.989\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#33CCCC\"><tspan>F</tspan></text>\n<text x=\"57.6038\" y=\"394\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C(C(F)(F)F)F</tspan></text>\n<path class=\"bond-0\" d=\"M 271.095,320.436 L 282.894,300\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 282.894,300 L 271.095,279.564\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 282.894,300 L 312.829,300\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 312.829,300 L 312.829,270.065\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 312.829,300 L 312.829,329.935\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 312.829,300 L 324.97,300\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 324.97,300 L 337.11,300\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"261.937\" y=\"331.413\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">2</tspan><tspan>H</tspan></text>\n<text x=\"261.937\" y=\"279.564\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">2</tspan><tspan>H</tspan></text>\n<text x=\"337.11\" y=\"304.989\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#00CC00\"><tspan>Cl</tspan></text>\n<text x=\"231.587\" y=\"394\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>[2H]C([2H])C(C)(C)Cl</tspan></text>\n<path class=\"bond-0\" d=\"M 432.42,306.444 L 459.648,294.005\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 459.648,294.005 L 484.035,311.365\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 484.035,311.365 L 511.263,298.926\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 511.263,298.926 L 535.65,316.287\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 511.263,298.926 L 512.446,286.521\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 512.446,286.521 L 513.629,274.115\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 535.65,316.287 L 548.432,310.447\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 548.432,310.447 L 561.213,304.608\" style=\"fill:none;fill-rule:evenodd;stroke:#A01EEF;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"561.213\" y=\"308.837\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#A01EEF\"><tspan>I</tspan></text>\n<text x=\"505.121\" y=\"274.115\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>OH</tspan></text>\n<text x=\"461.594\" y=\"394\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>CCCC(CI)O</tspan></text>\n<path class=\"bond-0\" d=\"M 609.091,304.949 L 635.423,290.711\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 635.423,290.711 L 660.92,306.396\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 660.92,306.396 L 687.252,292.157\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 687.252,292.157 L 712.748,307.843\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 712.748,307.843 L 739.08,293.604\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 739.08,293.604 L 758.504,305.553\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-6\" d=\"M 770.65,306.005 L 790.909,295.051\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"758.504\" y=\"314.777\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C</tspan><tspan style=\"baseline-shift:super;font-size:6.75px;\">+</tspan><tspan/></text>\n<text x=\"644.984\" y=\"394\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>CCCCCC[CH+]C</tspan></text>\n<path class=\"bond-0\" d=\"M 925.924,306.415 L 951.849,291.447\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 925.924,306.415 L 917.283,301.425\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 917.283,301.425 L 908.641,296.436\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 951.849,291.447 L 960.49,296.436\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 960.49,296.436 L 969.132,301.425\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 891.359,296.436 L 882.717,301.425\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 882.717,301.425 L 874.076,306.415\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 874.076,306.415 L 848.151,291.447\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 848.151,291.447 L 839.51,296.436\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 839.51,296.436 L 830.868,301.425\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"968.789\" y=\"311.404\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>OH</tspan></text>\n<text x=\"891.351\" y=\"296.436\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#0000FF\"><tspan>NH</tspan></text>\n<text x=\"813.243\" y=\"311.404\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>HO</tspan></text>\n<text x=\"857.993\" y=\"394\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C(CO)NCCO</tspan></text>\n<path class=\"bond-0\" d=\"M 170.127,493.823 L 145.106,510.255\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 145.106,510.255 L 118.364,496.802\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 118.364,496.802 L 93.3423,513.234\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 93.3423,513.234 L 81.969,507.512\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 81.969,507.512 L 70.5957,501.79\" style=\"fill:none;fill-rule:evenodd;stroke:#FF7F00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 62.6058,502.404 L 52.0924,509.308\" style=\"fill:none;fill-rule:evenodd;stroke:#FF7F00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 52.0924,509.308 L 41.579,516.212\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 66.3137,494.791 L 65.1968,475.383\" style=\"fill:none;fill-rule:evenodd;stroke:#FF7F00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"62.6058\" y=\"504.769\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF7F00\"><tspan>P</tspan></text>\n<text x=\"54.8967\" y=\"475.383\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF7F00\"><tspan>PH</tspan><tspan style=\"baseline-shift:sub;font-size:6.75px;\">2</tspan><tspan/></text>\n<text x=\"59.5918\" y=\"594\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>CCCCP(C)P</tspan></text>\n<path class=\"bond-0\" d=\"M 351.849,508.553 L 360.49,503.564\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-0\" d=\"M 360.49,503.564 L 369.132,498.575\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 351.849,508.553 L 341.217,502.414\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 341.217,502.414 L 330.584,496.276\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 321.265,496.276 L 304.66,505.863\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 295.34,505.863 L 278.735,496.276\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 269.416,496.276 L 258.783,502.414\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 258.783,502.414 L 248.151,508.553\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 248.151,508.553 L 239.51,503.564\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 239.51,503.564 L 230.868,498.575\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"368.789\" y=\"498.575\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>OH</tspan></text>\n<text x=\"321.265\" y=\"498.575\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>O</tspan></text>\n<text x=\"295.34\" y=\"513.542\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>O</tspan></text>\n<text x=\"269.416\" y=\"498.575\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>O</tspan></text>\n<text x=\"213.243\" y=\"498.575\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>HO</tspan></text>\n<text x=\"256.783\" y=\"594\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C(O)OOOCO</tspan></text>\n<path class=\"bond-0\" d=\"M 547.209,510.583 L 538.67,481.892\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 547.209,510.583 L 518.092,503.632\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 538.67,481.892 L 518.092,503.632\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 518.092,503.632 L 489.4,512.17\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 489.4,512.17 L 467.66,491.592\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 467.66,491.592 L 458.307,494.376\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 458.307,494.376 L 448.953,497.159\" style=\"fill:none;fill-rule:evenodd;stroke:#FF7F00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"428.984\" y=\"505.619\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF7F00\"><tspan>H</tspan><tspan style=\"baseline-shift:sub;font-size:6.75px;\">2</tspan><tspan>P</tspan></text>\n<text x=\"461.184\" y=\"594\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C1CC1CCP</tspan></text>\n<path class=\"bond-0\" d=\"M 770.127,493.823 L 759.946,500.509\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-0\" d=\"M 759.946,500.509 L 749.765,507.195\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 740.446,507.911 L 729.405,502.356\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 729.405,502.356 L 718.364,496.802\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 718.364,496.802 L 693.342,513.234\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 693.342,513.234 L 666.601,499.78\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 666.601,499.78 L 656.917,506.14\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 656.917,506.14 L 647.233,512.499\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 666.601,499.78 L 665.884,487.332\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 665.884,487.332 L 665.168,474.884\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"740.446\" y=\"515.244\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>O</tspan></text>\n<text x=\"635.925\" y=\"521.201\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#00CC00\"><tspan>Cl</tspan></text>\n<text x=\"659.227\" y=\"474.884\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#00CC00\"><tspan>Cl</tspan></text>\n<text x=\"655.198\" y=\"594\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>COCCC(Cl)Cl</tspan></text>\n<path class=\"bond-0\" d=\"M 873.976,528.211 L 892.204,514.967\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 892.204,514.967 L 920.674,524.218\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 898.325,510.661 L 918.254,517.136\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-6\" d=\"M 892.204,514.967 L 892.204,485.033\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 920.674,524.218 L 927.66,514.604\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 927.66,514.604 L 934.645,504.989\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 934.645,495.011 L 927.66,485.396\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 927.66,485.396 L 920.674,475.782\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 920.674,475.782 L 892.204,485.033\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 918.254,482.864 L 898.325,489.339\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 892.204,485.033 L 873.976,471.789\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"861.997\" y=\"538.051\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">2</tspan><tspan>H</tspan></text>\n<text x=\"934.275\" y=\"504.989\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#CCCC00\"><tspan>S</tspan></text>\n<text x=\"861.997\" y=\"472.925\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan style=\"baseline-shift:super;font-size:6.75px;\">2</tspan><tspan>H</tspan></text>\n<text x=\"831.551\" y=\"594\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>[2H]C1=CSC=C1[2H]</tspan></text>\n<path class=\"bond-0\" d=\"M 48.1511,705.987 L 74.0755,691.02\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 74.0755,691.02 L 100,705.987\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 100,705.987 L 125.924,691.02\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 125.924,691.02 L 136.06,696.871\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 136.06,696.871 L 146.195,702.723\" style=\"fill:none;fill-rule:evenodd;stroke:#00CC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"146.195\" y=\"710.976\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#00CC00\"><tspan>Cl</tspan></text>\n<text x=\"72.3986\" y=\"794\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>CCCCCl</tspan></text>\n<path class=\"bond-0\" d=\"M 285.942,712.375 L 275.31,706.236\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-0\" d=\"M 275.31,706.236 L 264.677,700.098\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-0\" d=\"M 282.949,717.56 L 272.316,711.421\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-0\" d=\"M 272.316,711.421 L 261.684,705.283\" style=\"fill:none;fill-rule:evenodd;stroke:#FF0000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 284.445,714.967 L 310.37,700\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 310.37,700 L 318.579,704.74\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 318.579,704.74 L 326.789,709.479\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 310.37,700 L 310.37,687.527\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 310.37,687.527 L 310.37,675.054\" style=\"fill:none;fill-rule:evenodd;stroke:#7F4C19;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"253.861\" y=\"704.989\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#FF0000\"><tspan>O</tspan></text>\n<text x=\"325.981\" y=\"720.456\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#0000FF\"><tspan>NH</tspan><tspan style=\"baseline-shift:sub;font-size:6.75px;\">2</tspan><tspan/></text>\n<text x=\"304.38\" y=\"675.054\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#7F4C19\"><tspan>Br</tspan></text>\n<text x=\"257.791\" y=\"794\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C(=O)C(N)Br</tspan></text>\n<path class=\"bond-0\" d=\"M 497.007,700 L 497.007,712.473\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-0\" d=\"M 497.007,712.473 L 497.007,724.946\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-0\" d=\"M 502.993,700 L 502.993,712.473\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-0\" d=\"M 502.993,712.473 L 502.993,724.946\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 500,700 L 491.197,694.918\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 491.197,694.918 L 482.395,689.836\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 500,700 L 510.466,693.957\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 510.466,693.957 L 520.932,687.915\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"496.005\" y=\"734.924\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#CCCC00\"><tspan>S</tspan></text>\n<text x=\"465.756\" y=\"690.022\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#CCCC00\"><tspan>HS</tspan></text>\n<text x=\"520.932\" y=\"690.521\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#CCCC00\"><tspan>S</tspan><tspan style=\"baseline-shift:super;font-size:6.75px;\">-</tspan><tspan/></text>\n<text x=\"460.182\" y=\"794\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C(=S)(S)[S-]</tspan></text>\n<path class=\"bond-0\" d=\"M 674.076,714.967 L 700,700\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 674.076,714.967 L 664.108,709.213\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 664.108,709.213 L 654.14,703.458\" style=\"fill:none;fill-rule:evenodd;stroke:#7F4C19;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 700,700 L 725.924,714.967\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 700,700 L 700,687.527\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 700,687.527 L 700,675.054\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 725.924,714.967 L 735.892,709.213\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 735.892,709.213 L 745.86,703.458\" style=\"fill:none;fill-rule:evenodd;stroke:#7F4C19;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"745.86\" y=\"704.989\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#7F4C19\"><tspan>Br</tspan></text>\n<text x=\"691.681\" y=\"675.054\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#CCCC00\"><tspan>SH</tspan></text>\n<text x=\"642.162\" y=\"704.989\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#7F4C19\"><tspan>Br</tspan></text>\n<text x=\"655.594\" y=\"794\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>C(C(CBr)S)Br</tspan></text>\n<path class=\"bond-0\" d=\"M 809.091,704.949 L 835.423,690.711\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 835.423,690.711 L 846.174,697.324\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-1\" d=\"M 846.174,697.324 L 856.925,703.938\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 864.915,704.236 L 876.083,698.197\" style=\"fill:none;fill-rule:evenodd;stroke:#CCCC00;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-2\" d=\"M 876.083,698.197 L 887.252,692.157\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-3\" d=\"M 887.252,692.157 L 912.748,707.843\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-4\" d=\"M 912.748,707.843 L 939.08,693.604\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-5\" d=\"M 939.08,693.604 L 964.577,709.289\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-6\" d=\"M 964.577,709.289 L 972.668,704.914\" style=\"fill:none;fill-rule:evenodd;stroke:#000000;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<path class=\"bond-6\" d=\"M 972.668,704.914 L 980.76,700.539\" style=\"fill:none;fill-rule:evenodd;stroke:#0000FF;stroke-width:2px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1\"/>\n<text x=\"856.925\" y=\"711.385\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#CCCC00\"><tspan>S</tspan></text>\n<text x=\"980.595\" y=\"700.539\" style=\"font-size:9px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#0000FF\"><tspan>NH</tspan><tspan style=\"baseline-shift:sub;font-size:6.75px;\">2</tspan><tspan/></text>\n<text x=\"858.792\" y=\"794\" style=\"font-size:12px;font-style:normal;font-weight:normal;fill-opacity:1;stroke:none;font-family:sans-serif;text-anchor:start;fill:#000000\"><tspan>CCSCCCCN</tspan></text>\n</svg>",
      "text/plain": [
       "<IPython.core.display.SVG object>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from rdkit.Chem import Draw\n",
    "\n",
    "number_to_print = 20\n",
    "mols = df['mol'][:number_to_print]\n",
    "\n",
    "#MolsToGridImage allows to show several molecules in a grid\n",
    "Draw.MolsToGridImage(mols, molsPerRow=5, useSVG=True, legends=list(df['smiles'][:number_to_print].values))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f1c768e6",
   "metadata": {},
   "source": [
    "## View it in 3D\n",
    "There's a nice Python package called [py3Dmol](https://3dmol.csb.pitt.edu/) which creates 3D images from the *Mol* object. This is not necessary, but it does create pretty pictures of our molecules that can be rotated and zoomed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "1ac479bb",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "209da5e974114a4eac15506c7591ca4e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(IntSlider(value=3064, description='Molecule', max=14609), Dropdown(description='Style:',…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import py3Dmol  # pip install py3Dmol\n",
    "from ipywidgets import interact,fixed,IntSlider\n",
    "import ipywidgets\n",
    "\n",
    "def show3D_molecule(idx, style):\n",
    "    \"\"\"\n",
    "    Show molecule in 3D\n",
    "    \"\"\"\n",
    "    mblock = Chem.MolToMolBlock(df['mol'].iloc[idx])\n",
    "    viewer = py3Dmol.view(width=300, height=300)\n",
    "    viewer.addModel(mblock, 'mol')\n",
    "    viewer.setStyle({style:{}})\n",
    "    viewer.rotate(45, \"y\", animationDuration=1)\n",
    "   \n",
    "    viewer.zoomTo()\n",
    "    \n",
    "    print(f\"SMILES notation: {df['smiles'].iloc[idx]}\\nRotate me!\");\n",
    "    \n",
    "    return viewer.show()\n",
    "  \n",
    "interact(show3D_molecule, \n",
    "         idx=ipywidgets.IntSlider(min=0,max=len(df[\"mol\"])-1, \n",
    "                                  step=1, value=3064, \n",
    "                                  description=\"Molecule\"),\n",
    "         style=ipywidgets.Dropdown(options=['line', 'stick', 'sphere'],\n",
    "                                   value='stick',\n",
    "                                   description='Style:'));"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b56d0f9e",
   "metadata": {},
   "source": [
    "## Use RD-Kit to create features for the model\n",
    "\n",
    "Now that we have the molecule in *Mol* format, we can use RD-Kit's built in functions to create new features for our model. For example, we can add columns for the number of total atoms in the molecule and the number **Chloride (Cl)** atoms in the molecule. These features might be important in predicting **logP**. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b063040",
   "metadata": {},
   "source": [
    "Note: The first function we use is `Chem.AddHs`. Typically, the SMILES format does not explicitly show the **Hydrogen atoms (H)**. They are implied. It actually helps us a bit in our model accuracy to specifically add these back. (Try commenting out the next line to see its effect on model performance.) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "37187a8e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# AddHs function adds H atoms to a MOL (as Hs in SMILES are usualy ignored)\n",
    "df['mol'] = df['mol'].apply(lambda x: Chem.AddHs(x))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6fd351df",
   "metadata": {},
   "source": [
    "## Additional features\n",
    "\n",
    "Let's add some other features such as the number of atoms in the molecule and the number of \"heavy\" atoms in the molecule. These might be important in determining the logP value. Look through the [RD-Kit documentation](https://www.rdkit.org/docs/source/rdkit.Chem.Lipinski.html) to add more features if you'd like to get a better model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "4fa3fc00",
   "metadata": {},
   "outputs": [],
   "source": [
    "# GetNumAtoms() method returns a general nubmer of all atoms in a molecule\n",
    "df['num_of_atoms'] = df['mol'].apply(lambda x: x.GetNumAtoms())\n",
    "\n",
    "# GetNumHeavyAtoms() method returns a nubmer of all atoms in a molecule with molecular weight > 1\n",
    "df['num_of_heavy_atoms'] = df['mol'].apply(lambda x: x.GetNumHeavyAtoms())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "e8712883",
   "metadata": {},
   "outputs": [],
   "source": [
    "#We're going to searches patterns and use it for a list of most common atoms only\n",
    "def number_of_atoms(atom_list, df):\n",
    "    for i in atom_list:\n",
    "        df['num_of_{}_atoms'.format(i)] = df['mol'].apply(lambda x: len(x.GetSubstructMatches(Chem.MolFromSmiles(i))))\n",
    "\n",
    "number_of_atoms(['C', 'O', 'N', 'Cl', 'P', 'Br', 'F'], df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "1a5393aa",
   "metadata": {},
   "outputs": [],
   "source": [
    "from rdkit.Chem import Descriptors\n",
    "df['tpsa'] = df['mol'].apply(lambda x: Descriptors.TPSA(x)) #https://en.wikipedia.org/wiki/Polar_surface_area\n",
    "df['mol_w'] = df['mol'].apply(lambda x: Descriptors.ExactMolWt(x)) # https://en.wikipedia.org/wiki/Molecular_mass\n",
    "df['num_valence_electrons'] = df['mol'].apply(lambda x: Descriptors.NumValenceElectrons(x)) # https://en.wikipedia.org/wiki/Valence_electron\n",
    "df['num_heteroatoms'] = df['mol'].apply(lambda x: Descriptors.NumHeteroatoms(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "31190607",
   "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>smiles</th>\n",
       "      <th>logP</th>\n",
       "      <th>mol</th>\n",
       "      <th>num_of_atoms</th>\n",
       "      <th>num_of_heavy_atoms</th>\n",
       "      <th>num_of_C_atoms</th>\n",
       "      <th>num_of_O_atoms</th>\n",
       "      <th>num_of_N_atoms</th>\n",
       "      <th>num_of_Cl_atoms</th>\n",
       "      <th>num_of_P_atoms</th>\n",
       "      <th>num_of_Br_atoms</th>\n",
       "      <th>num_of_F_atoms</th>\n",
       "      <th>tpsa</th>\n",
       "      <th>mol_w</th>\n",
       "      <th>num_valence_electrons</th>\n",
       "      <th>num_heteroatoms</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C[C@H]([C@@H](C)Cl)Cl</td>\n",
       "      <td>2.3</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc540fc9220&gt;</td>\n",
       "      <td>14</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>126.000306</td>\n",
       "      <td>38</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C(C=CBr)N</td>\n",
       "      <td>0.3</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc540fc9a00&gt;</td>\n",
       "      <td>11</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>26.02</td>\n",
       "      <td>134.968361</td>\n",
       "      <td>30</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CCC(CO)Br</td>\n",
       "      <td>1.3</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc540fc9a60&gt;</td>\n",
       "      <td>15</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>20.23</td>\n",
       "      <td>151.983677</td>\n",
       "      <td>38</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[13CH3][13CH2][13CH2][13CH2][13CH2][13CH2]O</td>\n",
       "      <td>2.0</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc540fc9ac0&gt;</td>\n",
       "      <td>21</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>20.23</td>\n",
       "      <td>108.124594</td>\n",
       "      <td>44</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CCCOCCP</td>\n",
       "      <td>0.6</td>\n",
       "      <td>&lt;rdkit.Chem.rdchem.Mol object at 0x7fc540fc9b20&gt;</td>\n",
       "      <td>20</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9.23</td>\n",
       "      <td>120.070402</td>\n",
       "      <td>44</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        smiles  logP  \\\n",
       "0                        C[C@H]([C@@H](C)Cl)Cl   2.3   \n",
       "1                                    C(C=CBr)N   0.3   \n",
       "2                                    CCC(CO)Br   1.3   \n",
       "3  [13CH3][13CH2][13CH2][13CH2][13CH2][13CH2]O   2.0   \n",
       "4                                      CCCOCCP   0.6   \n",
       "\n",
       "                                                mol  num_of_atoms  \\\n",
       "0  <rdkit.Chem.rdchem.Mol object at 0x7fc540fc9220>            14   \n",
       "1  <rdkit.Chem.rdchem.Mol object at 0x7fc540fc9a00>            11   \n",
       "2  <rdkit.Chem.rdchem.Mol object at 0x7fc540fc9a60>            15   \n",
       "3  <rdkit.Chem.rdchem.Mol object at 0x7fc540fc9ac0>            21   \n",
       "4  <rdkit.Chem.rdchem.Mol object at 0x7fc540fc9b20>            20   \n",
       "\n",
       "   num_of_heavy_atoms  num_of_C_atoms  num_of_O_atoms  num_of_N_atoms  \\\n",
       "0                   6               4               0               0   \n",
       "1                   5               3               0               1   \n",
       "2                   6               4               1               0   \n",
       "3                   7               6               1               0   \n",
       "4                   7               5               1               0   \n",
       "\n",
       "   num_of_Cl_atoms  num_of_P_atoms  num_of_Br_atoms  num_of_F_atoms   tpsa  \\\n",
       "0                2               0                0               0   0.00   \n",
       "1                0               0                1               0  26.02   \n",
       "2                0               0                1               0  20.23   \n",
       "3                0               0                0               0  20.23   \n",
       "4                0               1                0               0   9.23   \n",
       "\n",
       "        mol_w  num_valence_electrons  num_heteroatoms  \n",
       "0  126.000306                     38                2  \n",
       "1  134.968361                     30                2  \n",
       "2  151.983677                     38                2  \n",
       "3  108.124594                     44                1  \n",
       "4  120.070402                     44                2  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ef29b9d",
   "metadata": {},
   "source": [
    "## Use the GenSim word2vec model\n",
    "\n",
    "[GenSim](https://radimrehurek.com/gensim/index.html#) is a natural language processing ([NLP](https://en.wikipedia.org/wiki/Natural_language_processing)) Python library for creating topic modeling and word embedding ([word vectors](https://en.wikipedia.org/wiki/Word2vec)). A word vector contains a list of floating point numbers that uniquely identify the word (see figure below). Remember that computers only really understand numbers (not language) so we need to convert the SMILES notation to a vector of numbers in order to use it in machine learning. If we do it correctly, then vectors that are \"close\" together in this word space are typically similar in meaning. More information on word embeddings can be found [here](https://en.wikipedia.org/wiki/Word_embedding).\n",
    "\n",
    "For our model, we'll consider each molecule to be a \"word\" in the language of chemistry. So **C\\[C@H\\](\\[C@@H\\](C)Cl)Cl** could replace **Houses** in the figure below and **C(C=CBr)N** could replace **puppy**. We can then create a word vector embedding for our molecule (the numbers in **d1-d7**). In this case, we'll use vectors of length 300 and vectors that are close together represent molecules which should have similar chemical properties (including logP if we are lucky).\n",
    "\n",
    "<br>\n",
    "<br>\n",
    "\n",
    "![word embeddings](https://journals.plos.org/plosone/article/figure/image?size=large&download=&id=10.1371/journal.pone.0231189.g008)\n",
    "\n",
    "Figure citation: Rozado D (2020) Wide range screening of algorithmic bias in word embedding models using large sentiment lexicons reveals underreported bias types. PLoS ONE 15(4): e0231189. https://doi.org/10.1371/journal.pone.0231189, Creative commons license"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5916093a",
   "metadata": {},
   "source": [
    "## mol2vec\n",
    "\n",
    "[Mol2Vec](https://pubs.acs.org/doi/10.1021/acs.jcim.7b00616) is an unsupervised machine learning approach to learn vector representations (word embeddings) of molecular substructures. We've already converted the SMILES notation of the chemical into a Mol Python object. Mol2Vec converts that object into a word2vec vector (a sentence) that we will use as an additional feature of the model."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3db2c9d1",
   "metadata": {},
   "source": [
    "## Installing GenSim and Mol2Vec \n",
    "\n",
    "`pip install git+https://github.com/samoturk/mol2vec` which will get [GenSim](https://radimrehurek.com/gensim/index.html) and [mol2vec](https://github.com/samoturk/mol2vec).\n",
    "\n",
    "Note: mol2vec currently has an error with GenSim 4. \n",
    "\n",
    ">AttributeError: The vocab attribute was removed from KeyedVector in Gensim 4.0.0.\n",
    "Use KeyedVector's .key_to_index dict, .index_to_key list, and methods .get_vecattr(key, attr) and .set_vecattr(key, attr, new_val) instead.\n",
    "See https://github.com/RaRe-Technologies/gensim/wiki/Migrating-from-Gensim-3.x-to-4\n",
    "\n",
    "I have created a [patch](https://github.com/samoturk/mol2vec/pull/16) here which you can just use to overwrite the file on your system. (Hopefully, the PR will be merged and this step won't be needed.) On my system, the `features.py` file is `~/anaconda3/envs/chem/lib/python3.8/site-packages/mol2vec/features.py` so I did.\n",
    "\n",
    "`wget https://raw.githubusercontent.com/tonyreina/mol2vec/master/mol2vec/features.py -O ~/anaconda3/envs/chem/lib/python3.8/site-packages/mol2vec/features.py`\n",
    "\n",
    "You may need to restart the Jupyter Notebook server for the change to take effect.\n",
    "\n",
    "You should also download the pre-trained [model_300dim.pkl](https://github.com/samoturk/mol2vec_notebooks/blob/master/Notebooks/model_300dim.pkl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "309b1167",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "26567327"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import requests\n",
    "\n",
    "url = 'https://github.com/samoturk/mol2vec_notebooks/raw/master/Notebooks/model_300dim.pkl'\n",
    "file_to_download = requests.get(url, allow_redirects=True)\n",
    "\n",
    "open('model_300dim.pkl', 'wb').write(file_to_download.content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "c821376a",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/reina/anaconda3/envs/chem/lib/python3.8/site-packages/gensim/similarities/__init__.py:15: UserWarning: The gensim.similarities.levenshtein submodule is disabled, because the optional Levenshtein package <https://pypi.org/project/python-Levenshtein/> is unavailable. Install Levenhstein (e.g. `pip install python-Levenshtein`) to suppress this warning.\n",
      "  warnings.warn(msg)\n"
     ]
    }
   ],
   "source": [
    "from mol2vec.features import mol2alt_sentence, mol2sentence, MolSentence, DfVec, sentences2vec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "65b42caf",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Loading pre-trained model via word2vec\n",
    "from gensim.models import word2vec  # https://radimrehurek.com/gensim/index.html#install\n",
    "w2vec_model = word2vec.Word2Vec.load('model_300dim.pkl') # Download from https://github.com/samoturk/mol2vec_notebooks/blob/master/Notebooks/model_300dim.pkl"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f888980b",
   "metadata": {},
   "source": [
    "## Mol2Vec Sentences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "eabfda09",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Constructing sentences\n",
    "df['sentence'] = df.apply(lambda x: MolSentence(mol2alt_sentence(x['mol'], 1)), axis=1)\n",
    "\n",
    "#Extracting embeddings to a numpy.array\n",
    "#Note that we always should mark unseen='UNK' in sentence2vec() so that model is taught how to handle unknown substructures\n",
    "df['mol2vec'] = [DfVec(x) for x in sentences2vec(df['sentence'], w2vec_model, unseen='UNK')]\n",
    "\n",
    "y = df['logP'].values\n",
    "X = np.array([x.vec for x in df['mol2vec']])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30d391b1",
   "metadata": {},
   "source": [
    "## Combine features\n",
    "\n",
    "Here we'll combine the Mol2Vec Sentences (a chemical word2vec vector) with the RD-Kit calculated features (such as, number of Oxygen atoms in the molecule). We'll use these to predict the logP value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "acf70010",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = pd.concat((pd.DataFrame(X), df.drop([\"mol2vec\", \"sentence\", \"smiles\", \"mol\", \"logP\"], axis=1)), axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0cf0607c",
   "metadata": {},
   "source": [
    "## Train/Validation/Test Splits\n",
    "\n",
    "Finally, we split the data into random training, validation, and testing splits as we would for any machine learning project.\n",
    "\n",
    "`!pip install scikit-learn`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "f8ca22ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import RidgeCV   #!pip install scikit-learn\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import mean_absolute_error, mean_squared_error\n",
    "\n",
    "X_train, X_remain, y_train, y_remain = train_test_split(X, y, test_size=.2, random_state=816)\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_remain, y_remain, test_size=.5, random_state=817)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a13fc118",
   "metadata": {},
   "source": [
    "## Ridge regression\n",
    "\n",
    "Let's first try a [ridge regression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.ridge_regression.html?highlight=regression#sklearn.linear_model.ridge_regression) model using Scikit Learn to predict the [logP](https://en.wikipedia.org/wiki/Partition_coefficient_) from the [Mol2Vec](https://github.com/samoturk/mol2vec).\n",
    "\n",
    "Feel free to try other classical machine learning regression models if you'd like."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f961acd0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RidgeCV(alphas=array([ 0.1,  1. , 10. ]))"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ridge_model = RidgeCV()\n",
    "ridge_model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "449e2d22",
   "metadata": {},
   "source": [
    "## Plot the performance of the model\n",
    "\n",
    "We'll plot the accuracy of the model predictions using the function below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "2de22a2a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_evaluation(model, X_test, y_test, title=\"Test\"):\n",
    "    \"\"\"\n",
    "    Scatter plot between target and prediction logP values\n",
    "    \"\"\"\n",
    "    \n",
    "    prediction = np.squeeze(model.predict(X_test))\n",
    "    mae = mean_absolute_error(y_test, prediction)\n",
    "    mse = mean_squared_error(y_test, prediction)\n",
    "\n",
    "    plt.figure(figsize=(10,10))\n",
    "    sns.regplot(x=prediction, y=y_test, scatter_kws={\"color\": \"blue\"}, line_kws={\"color\": \"red\"});\n",
    "    plt.xlabel('Model Prediction', fontsize=24);\n",
    "    plt.ylabel('Measured logP', fontsize=24);\n",
    "    plt.xticks(fontsize=24);\n",
    "    plt.yticks(fontsize=24);\n",
    "    plt.suptitle('Predicting logP from SMILES', fontsize=32);\n",
    "    plt.title(f'{title}: MAE = {mae:.3f}, MSE = {mse:.3f}', fontsize=24);\n",
    "    plt.grid(True);\n",
    "         \n",
    "    cc = np.corrcoef(y_test, prediction)\n",
    "    plt.text(-1.5, 2, f\"Correlation\\ncoefficient\\n{cc[0][1]:.4f}\", fontsize=24, \n",
    "             horizontalalignment=\"center\", backgroundcolor=\"black\", color=\"white\")\n",
    "#     print('MAE score:', round(mae, 4))\n",
    "#     print('MSE score:', round(mse,4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "22b75c1a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_evaluation(ridge_model, X_test, y_test, \"Ridge model\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "20fe88f4",
   "metadata": {},
   "source": [
    "## Plot some example predictions\n",
    "\n",
    "Let's plot the molecular structure and measured/predicted logP values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "a371bb86",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_example(model, X_test, y_test, df, idx=0, show_3d = True):\n",
    "    \"\"\"\n",
    "    Plot the chemical, the measured logP, and the predicted logP\n",
    "    \"\"\"\n",
    "\n",
    "    prediction = np.squeeze(model.predict(np.expand_dims(X_test.iloc[idx], 0))[0])\n",
    "    \n",
    "    target = df[df.index==X_test.index[idx]][\"logP\"].values[0]\n",
    "    \n",
    "    if show_3d:\n",
    "        show3D_molecule(idx, \"stick\")\n",
    "    else:\n",
    "        display(df[df.index==X_test.index[idx]][\"mol\"].values[0])\n",
    "    \n",
    "    smiles = df[df.index==X_test.index[idx]][\"smiles\"].values[0]\n",
    "    \n",
    "    print(f\"SMILES notation: {smiles}\")\n",
    "    print(f\"Target logP = {target:.3f}, Prediction logP = {prediction:.3f}\")\n",
    "    print(\"=\"*50)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "2670c1e5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CN(C)C(N)N\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224830678540142\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224830678540142\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224830678540142 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224830678540142\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224830678540142 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678540142\"),{backgroundColor:\"white\"});\nviewer_16224830678540142.zoomTo();\n\tviewer_16224830678540142.addModel(\"\\n     RDKit          2D\\n\\n 17 16  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500   -1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.5490    0.5490    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.9510   -2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0653   -0.9530    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.9510    2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.8481    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.6462   -0.2422    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  2  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  3 10  1  0\\n  3 11  1  0\\n  3 12  1  0\\n  4 13  1  0\\n  5 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\n  6 17  1  0\\nM  END\\n\",\"mol\");\n\tviewer_16224830678540142.setStyle({\"stick\": {}});\n\tviewer_16224830678540142.rotate(45,\"y\");\n\tviewer_16224830678540142.zoomTo();\nviewer_16224830678540142.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224830678540142\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224830678540142\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224830678540142 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224830678540142\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224830678540142 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678540142\"),{backgroundColor:\"white\"});\n",
       "viewer_16224830678540142.zoomTo();\n",
       "\tviewer_16224830678540142.addModel(\"\\n     RDKit          2D\\n\\n 17 16  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500   -1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.5490    0.5490    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.9510   -2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0653   -0.9530    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.9510    2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.8481    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.6462   -0.2422    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  2  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  3 10  1  0\\n  3 11  1  0\\n  3 12  1  0\\n  4 13  1  0\\n  5 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\n  6 17  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224830678540142.setStyle({\"stick\": {}});\n",
       "\tviewer_16224830678540142.rotate(45,\"y\");\n",
       "\tviewer_16224830678540142.zoomTo();\n",
       "viewer_16224830678540142.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: C1=NN=[C+]S1\n",
      "Target logP = 0.400, Prediction logP = 0.258\n",
      "==================================================\n",
      "SMILES notation: C(C(O)ON)O\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224830678574045\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224830678574045\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224830678574045 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224830678574045\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224830678574045 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678574045\"),{backgroundColor:\"white\"});\nviewer_16224830678574045.zoomTo();\n\tviewer_16224830678574045.addModel(\"\\n     RDKit          2D\\n\\n 13 12  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.7990   -2.2500    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.7990   -3.7500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.0981   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -2.2500   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  2  4  1  0\\n  4  5  1  0\\n  1  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  2  9  1  0\\n  3 10  1  0\\n  5 11  1  0\\n  5 12  1  0\\n  6 13  1  0\\nM  END\\n\",\"mol\");\n\tviewer_16224830678574045.setStyle({\"stick\": {}});\n\tviewer_16224830678574045.rotate(45,\"y\");\n\tviewer_16224830678574045.zoomTo();\nviewer_16224830678574045.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224830678574045\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224830678574045\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224830678574045 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224830678574045\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224830678574045 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678574045\"),{backgroundColor:\"white\"});\n",
       "viewer_16224830678574045.zoomTo();\n",
       "\tviewer_16224830678574045.addModel(\"\\n     RDKit          2D\\n\\n 13 12  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.7990   -2.2500    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.7990   -3.7500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.0981   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -2.2500   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  2  4  1  0\\n  4  5  1  0\\n  1  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  2  9  1  0\\n  3 10  1  0\\n  5 11  1  0\\n  5 12  1  0\\n  6 13  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224830678574045.setStyle({\"stick\": {}});\n",
       "\tviewer_16224830678574045.rotate(45,\"y\");\n",
       "\tviewer_16224830678574045.zoomTo();\n",
       "viewer_16224830678574045.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: COCCCNN\n",
      "Target logP = -0.700, Prediction logP = -0.852\n",
      "==================================================\n",
      "SMILES notation: C(=O)(N)I\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_1622483067862554\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_1622483067862554\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_1622483067862554 = null;\nvar warn = document.getElementById(\"3dmolwarning_1622483067862554\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_1622483067862554 = $3Dmol.createViewer($(\"#3dmolviewer_1622483067862554\"),{backgroundColor:\"white\"});\nviewer_1622483067862554.zoomTo();\n\tviewer_1622483067862554.addModel(\"\\n     RDKit          2D\\n\\n  6  5  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 I   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.5981   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  2  0\\n  1  3  1  0\\n  1  4  1  0\\n  3  5  1  0\\n  3  6  1  0\\nM  END\\n\",\"mol\");\n\tviewer_1622483067862554.setStyle({\"stick\": {}});\n\tviewer_1622483067862554.rotate(45,\"y\");\n\tviewer_1622483067862554.zoomTo();\nviewer_1622483067862554.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_1622483067862554\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_1622483067862554\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_1622483067862554 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_1622483067862554\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_1622483067862554 = $3Dmol.createViewer($(\"#3dmolviewer_1622483067862554\"),{backgroundColor:\"white\"});\n",
       "viewer_1622483067862554.zoomTo();\n",
       "\tviewer_1622483067862554.addModel(\"\\n     RDKit          2D\\n\\n  6  5  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 I   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.5981   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  2  0\\n  1  3  1  0\\n  1  4  1  0\\n  3  5  1  0\\n  3  6  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_1622483067862554.setStyle({\"stick\": {}});\n",
       "\tviewer_1622483067862554.rotate(45,\"y\");\n",
       "\tviewer_1622483067862554.zoomTo();\n",
       "viewer_1622483067862554.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CS(=O)P\n",
      "Target logP = -0.600, Prediction logP = -0.450\n",
      "==================================================\n",
      "SMILES notation: CCCCOCC\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_1622483067865381\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_1622483067865381\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_1622483067865381 = null;\nvar warn = document.getElementById(\"3dmolwarning_1622483067865381\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_1622483067865381 = $3Dmol.createViewer($(\"#3dmolviewer_1622483067865381\"),{backgroundColor:\"white\"});\nviewer_1622483067865381.zoomTo();\n\tviewer_1622483067865381.addModel(\"\\n     RDKit          2D\\n\\n 21 20  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.7500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    2.5981    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.4510    2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.0490    0.5490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.2010    3.3481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.7990    1.8481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  3 14  1  0\\n  4 15  1  0\\n  4 16  1  0\\n  6 17  1  0\\n  6 18  1  0\\n  7 19  1  0\\n  7 20  1  0\\n  7 21  1  0\\nM  END\\n\",\"mol\");\n\tviewer_1622483067865381.setStyle({\"stick\": {}});\n\tviewer_1622483067865381.rotate(45,\"y\");\n\tviewer_1622483067865381.zoomTo();\nviewer_1622483067865381.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_1622483067865381\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_1622483067865381\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_1622483067865381 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_1622483067865381\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_1622483067865381 = $3Dmol.createViewer($(\"#3dmolviewer_1622483067865381\"),{backgroundColor:\"white\"});\n",
       "viewer_1622483067865381.zoomTo();\n",
       "\tviewer_1622483067865381.addModel(\"\\n     RDKit          2D\\n\\n 21 20  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.7500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    2.5981    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.4510    2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.0490    0.5490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.2010    3.3481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.7990    1.8481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  3 14  1  0\\n  4 15  1  0\\n  4 16  1  0\\n  6 17  1  0\\n  6 18  1  0\\n  7 19  1  0\\n  7 20  1  0\\n  7 21  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_1622483067865381.setStyle({\"stick\": {}});\n",
       "\tviewer_1622483067865381.rotate(45,\"y\");\n",
       "\tviewer_1622483067865381.zoomTo();\n",
       "viewer_1622483067865381.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CP1[SH2]C[SH2]1\n",
      "Target logP = 0.700, Prediction logP = 0.954\n",
      "==================================================\n",
      "SMILES notation: [C](F)(F)Br\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224830678699508\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224830678699508\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224830678699508 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224830678699508\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224830678699508 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678699508\"),{backgroundColor:\"white\"});\nviewer_16224830678699508.zoomTo();\n\tviewer_16224830678699508.addModel(\"\\n     RDKit          2D\\n\\n  4  3  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  3  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 F   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 F   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 Br  0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  1  3  1  0\\n  1  4  1  0\\nM  RAD  1   1   2\\nM  END\\n\",\"mol\");\n\tviewer_16224830678699508.setStyle({\"stick\": {}});\n\tviewer_16224830678699508.rotate(45,\"y\");\n\tviewer_16224830678699508.zoomTo();\nviewer_16224830678699508.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224830678699508\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224830678699508\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224830678699508 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224830678699508\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224830678699508 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678699508\"),{backgroundColor:\"white\"});\n",
       "viewer_16224830678699508.zoomTo();\n",
       "\tviewer_16224830678699508.addModel(\"\\n     RDKit          2D\\n\\n  4  3  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  3  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 F   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 F   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 Br  0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  1  3  1  0\\n  1  4  1  0\\nM  RAD  1   1   2\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224830678699508.setStyle({\"stick\": {}});\n",
       "\tviewer_16224830678699508.rotate(45,\"y\");\n",
       "\tviewer_16224830678699508.zoomTo();\n",
       "viewer_16224830678699508.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CNSSNC\n",
      "Target logP = 0.400, Prediction logP = 0.351\n",
      "==================================================\n",
      "SMILES notation: C(C[NH3+])C(O)O\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224830678737242\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224830678737242\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224830678737242 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224830678737242\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224830678737242 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678737242\"),{backgroundColor:\"white\"});\nviewer_16224830678737242.zoomTo();\n\tviewer_16224830678737242.addModel(\"\\n     RDKit          2D\\n\\n 16 15  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 N   0  0  0  0  0  4  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -3.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000   -1.5000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -3.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -2.7990   -2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  1  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  2  9  1  0\\n  2 10  1  0\\n  3 11  1  0\\n  3 12  1  0\\n  3 13  1  0\\n  4 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\nM  CHG  1   3   1\\nM  END\\n\",\"mol\");\n\tviewer_16224830678737242.setStyle({\"stick\": {}});\n\tviewer_16224830678737242.rotate(45,\"y\");\n\tviewer_16224830678737242.zoomTo();\nviewer_16224830678737242.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224830678737242\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224830678737242\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224830678737242 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224830678737242\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224830678737242 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678737242\"),{backgroundColor:\"white\"});\n",
       "viewer_16224830678737242.zoomTo();\n",
       "\tviewer_16224830678737242.addModel(\"\\n     RDKit          2D\\n\\n 16 15  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 N   0  0  0  0  0  4  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -3.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000   -1.5000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -3.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -2.7990   -2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  1  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  2  9  1  0\\n  2 10  1  0\\n  3 11  1  0\\n  3 12  1  0\\n  3 13  1  0\\n  4 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\nM  CHG  1   3   1\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224830678737242.setStyle({\"stick\": {}});\n",
       "\tviewer_16224830678737242.rotate(45,\"y\");\n",
       "\tviewer_16224830678737242.zoomTo();\n",
       "viewer_16224830678737242.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CC([CH2])(Cl)Cl\n",
      "Target logP = 2.100, Prediction logP = 2.488\n",
      "==================================================\n",
      "SMILES notation: C=CCCCl\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_1622483067876679\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_1622483067876679\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_1622483067876679 = null;\nvar warn = document.getElementById(\"3dmolwarning_1622483067876679\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_1622483067876679 = $3Dmol.createViewer($(\"#3dmolviewer_1622483067876679\"),{backgroundColor:\"white\"});\nviewer_1622483067876679.zoomTo();\n\tviewer_1622483067876679.addModel(\"\\n     RDKit          2D\\n\\n 12 11  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.5981   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.8971   -0.7500    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.1962   -1.5000    0.0000 Cl  0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.3481    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.8481   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.1471   -2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.6471    0.5490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  2  3\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  1  6  1  0\\n  1  7  1  0\\n  2  8  1  0\\n  3  9  1  0\\n  3 10  1  0\\n  4 11  1  0\\n  4 12  1  0\\nM  END\\n\",\"mol\");\n\tviewer_1622483067876679.setStyle({\"stick\": {}});\n\tviewer_1622483067876679.rotate(45,\"y\");\n\tviewer_1622483067876679.zoomTo();\nviewer_1622483067876679.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_1622483067876679\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_1622483067876679\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_1622483067876679 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_1622483067876679\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_1622483067876679 = $3Dmol.createViewer($(\"#3dmolviewer_1622483067876679\"),{backgroundColor:\"white\"});\n",
       "viewer_1622483067876679.zoomTo();\n",
       "\tviewer_1622483067876679.addModel(\"\\n     RDKit          2D\\n\\n 12 11  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.5981   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.8971   -0.7500    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.1962   -1.5000    0.0000 Cl  0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.3481    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.8481   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.1471   -2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.6471    0.5490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  2  3\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  1  6  1  0\\n  1  7  1  0\\n  2  8  1  0\\n  3  9  1  0\\n  3 10  1  0\\n  4 11  1  0\\n  4 12  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_1622483067876679.setStyle({\"stick\": {}});\n",
       "\tviewer_1622483067876679.rotate(45,\"y\");\n",
       "\tviewer_1622483067876679.zoomTo();\n",
       "viewer_1622483067876679.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: [2H]C([2H])([2H])C([2H])(CO)C([2H])([2H])[2H]\n",
      "Target logP = 0.800, Prediction logP = 0.815\n",
      "==================================================\n",
      "SMILES notation: CCO[S+](C)C\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224830678817647\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224830678817647\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224830678817647 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224830678817647\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224830678817647 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678817647\"),{backgroundColor:\"white\"});\nviewer_16224830678817647.zoomTo();\n\tviewer_16224830678817647.addModel(\"\\n     RDKit          2D\\n\\n 17 16  0  0  0  0  0  0  0  0999 V2000\\n    0.0000   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 S   0  0  0  0  0  3  0  0  0  0  0  0\\n    5.2500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.3519    1.2059    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    2.7990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500   -0.2010    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.8816    2.1234    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.2990    3.3481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  2 10  1  0\\n  2 11  1  0\\n  5 12  1  0\\n  5 13  1  0\\n  5 14  1  0\\n  6 15  1  0\\n  6 16  1  0\\n  6 17  1  0\\nM  CHG  1   4   1\\nM  END\\n\",\"mol\");\n\tviewer_16224830678817647.setStyle({\"stick\": {}});\n\tviewer_16224830678817647.rotate(45,\"y\");\n\tviewer_16224830678817647.zoomTo();\nviewer_16224830678817647.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224830678817647\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224830678817647\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224830678817647 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224830678817647\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224830678817647 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678817647\"),{backgroundColor:\"white\"});\n",
       "viewer_16224830678817647.zoomTo();\n",
       "\tviewer_16224830678817647.addModel(\"\\n     RDKit          2D\\n\\n 17 16  0  0  0  0  0  0  0  0999 V2000\\n    0.0000   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 S   0  0  0  0  0  3  0  0  0  0  0  0\\n    5.2500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.3519    1.2059    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    2.7990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500   -0.2010    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.8816    2.1234    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.2990    3.3481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  2 10  1  0\\n  2 11  1  0\\n  5 12  1  0\\n  5 13  1  0\\n  5 14  1  0\\n  6 15  1  0\\n  6 16  1  0\\n  6 17  1  0\\nM  CHG  1   4   1\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224830678817647.setStyle({\"stick\": {}});\n",
       "\tviewer_16224830678817647.rotate(45,\"y\");\n",
       "\tviewer_16224830678817647.zoomTo();\n",
       "viewer_16224830678817647.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CNC(CS)N\n",
      "Target logP = -0.700, Prediction logP = -0.346\n",
      "==================================================\n",
      "SMILES notation: [13CH3][13CH2][13CH2][13CH2][13CH2][13CH2]O\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224830678862119\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224830678862119\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224830678862119 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224830678862119\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224830678862119 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678862119\"),{backgroundColor:\"white\"});\nviewer_16224830678862119.zoomTo();\n\tviewer_16224830678862119.addModel(\"\\n     RDKit          2D\\n\\n 21 20  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    9.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    9.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  3 14  1  0\\n  4 15  1  0\\n  4 16  1  0\\n  5 17  1  0\\n  5 18  1  0\\n  6 19  1  0\\n  6 20  1  0\\n  7 21  1  0\\nM  ISO  6   1  13   2  13   3  13   4  13   5  13   6  13\\nM  END\\n\",\"mol\");\n\tviewer_16224830678862119.setStyle({\"stick\": {}});\n\tviewer_16224830678862119.rotate(45,\"y\");\n\tviewer_16224830678862119.zoomTo();\nviewer_16224830678862119.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224830678862119\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224830678862119\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224830678862119 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224830678862119\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224830678862119 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678862119\"),{backgroundColor:\"white\"});\n",
       "viewer_16224830678862119.zoomTo();\n",
       "\tviewer_16224830678862119.addModel(\"\\n     RDKit          2D\\n\\n 21 20  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    9.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    9.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  3 14  1  0\\n  4 15  1  0\\n  4 16  1  0\\n  5 17  1  0\\n  5 18  1  0\\n  6 19  1  0\\n  6 20  1  0\\n  7 21  1  0\\nM  ISO  6   1  13   2  13   3  13   4  13   5  13   6  13\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224830678862119.setStyle({\"stick\": {}});\n",
       "\tviewer_16224830678862119.rotate(45,\"y\");\n",
       "\tviewer_16224830678862119.zoomTo();\n",
       "viewer_16224830678862119.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: COCC(O)I\n",
      "Target logP = 0.200, Prediction logP = 0.561\n",
      "==================================================\n",
      "SMILES notation: CCNNNNI\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224830678903794\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224830678903794\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224830678903794 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224830678903794\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224830678903794 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678903794\"),{backgroundColor:\"white\"});\nviewer_16224830678903794.zoomTo();\n\tviewer_16224830678903794.addModel(\"\\n     RDKit          2D\\n\\n 16 15  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    1.2990    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    2.5981    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    2.5981    0.0000 I   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  4 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\nM  END\\n\",\"mol\");\n\tviewer_16224830678903794.setStyle({\"stick\": {}});\n\tviewer_16224830678903794.rotate(45,\"y\");\n\tviewer_16224830678903794.zoomTo();\nviewer_16224830678903794.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224830678903794\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224830678903794\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224830678903794 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224830678903794\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224830678903794 = $3Dmol.createViewer($(\"#3dmolviewer_16224830678903794\"),{backgroundColor:\"white\"});\n",
       "viewer_16224830678903794.zoomTo();\n",
       "\tviewer_16224830678903794.addModel(\"\\n     RDKit          2D\\n\\n 16 15  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    1.2990    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    2.5981    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    2.5981    0.0000 I   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  4 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224830678903794.setStyle({\"stick\": {}});\n",
       "\tviewer_16224830678903794.rotate(45,\"y\");\n",
       "\tviewer_16224830678903794.zoomTo();\n",
       "viewer_16224830678903794.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: [CH2]CC(CO)N\n",
      "Target logP = -0.400, Prediction logP = -0.329\n",
      "==================================================\n"
     ]
    }
   ],
   "source": [
    "indicies = np.random.choice(range(len(X_test)), 10) # Try a few examples\n",
    "\n",
    "for idx in indicies:\n",
    "    plot_example(ridge_model, X_test, y_test, df, idx, show_3d=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52225326",
   "metadata": {},
   "source": [
    "## TensorFlow Model\n",
    "\n",
    "Let's create a simple multi-layer neural network using TensorFlow. We can compare our TensorFlow neural network with the Ridge Regression model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "01eb25c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b73084a",
   "metadata": {},
   "source": [
    "## A very simple multi-layered perceptron with dropout\n",
    "\n",
    "This is about as simple a neural network topology as you can get."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "54180251",
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = tf.keras.layers.Input(shape=(X_train.shape[1]), name=\"mol2vec\")\n",
    "\n",
    "dense1 = tf.keras.layers.Dense(128, activation=\"relu\")(inputs)\n",
    "drop1  = tf.keras.layers.Dropout(0.3)(dense1)\n",
    "prediction = tf.keras.layers.Dense(1, activation=\"linear\", name=\"logP_value\")(drop1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "36cb3fa3",
   "metadata": {},
   "outputs": [],
   "source": [
    "tf_model = tf.keras.models.Model(inputs=[inputs], outputs=[prediction], name=\"logP_Model\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8c8e4d87",
   "metadata": {},
   "source": [
    "## Print TensorFlow model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c4f992c1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"logP_Model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "mol2vec (InputLayer)         [(None, 113)]             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 128)               14592     \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "logP_value (Dense)           (None, 1)                 129       \n",
      "=================================================================\n",
      "Total params: 14,721\n",
      "Trainable params: 14,721\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "tf_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8571069f",
   "metadata": {},
   "source": [
    "## Loss and metrics\n",
    "\n",
    "We'll use the mean average error ([MAE](https://en.wikipedia.org/wiki/Mean_absolute_error) or L1-norm) and mean squared error ([MSE](https://en.wikipedia.org/wiki/Mean_squared_error) or L2-norm) for the loss and goodness of fit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "ee57fed7",
   "metadata": {},
   "outputs": [],
   "source": [
    "tf_model.compile(loss=\"mse\",\n",
    "              metrics=[\"mse\", \"mae\"],\n",
    "              optimizer=tf.keras.optimizers.Adam(lr=3e-4)) # Learning rate must be low"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc38da89",
   "metadata": {},
   "source": [
    "## Add callbacks\n",
    "\n",
    "We'll add early stopping, TensorBoard, and model checkpoints.\n",
    "\n",
    "You can view the TensorBoard by running:\n",
    "\n",
    "`tensorboard --logdir logP_model`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "bdae37ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext tensorboard\n",
    "\n",
    "stopping = tf.keras.callbacks.EarlyStopping(patience=30, verbose=1, restore_best_weights=True)\n",
    "\n",
    "tensorboard = tf.keras.callbacks.TensorBoard(log_dir='logs')\n",
    "\n",
    "checkpoint_dir = \"logP_model\"\n",
    "checkpoint = tf.keras.callbacks.ModelCheckpoint(checkpoint_dir, monitor='val_loss', verbose=1, save_best_only=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "958f35f6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Reusing TensorBoard on port 6006 (pid 29385), started 1 day, 15:36:25 ago. (Use '!kill 29385' to kill it.)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "      <iframe id=\"tensorboard-frame-6b54bb5a13d5772\" width=\"100%\" height=\"800\" frameborder=\"0\">\n",
       "      </iframe>\n",
       "      <script>\n",
       "        (function() {\n",
       "          const frame = document.getElementById(\"tensorboard-frame-6b54bb5a13d5772\");\n",
       "          const url = new URL(\"/\", window.location);\n",
       "          const port = 6006;\n",
       "          if (port) {\n",
       "            url.port = port;\n",
       "          }\n",
       "          frame.src = url;\n",
       "        })();\n",
       "      </script>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%tensorboard --logdir logs"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0afc5be9",
   "metadata": {},
   "source": [
    "## Train the TensorFlow model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e4cce222",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "tf_model.fit(X_train, y_train, validation_data=(X_val, y_val), epochs=1000, callbacks=[stopping, tensorboard, checkpoint])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81e43137",
   "metadata": {},
   "source": [
    "## Test the accuracy of the final TensorFlow model on the holdout test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "707a591a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "46/46 [==============================] - 0s 670us/step - loss: 0.1244 - mse: 0.1244 - mae: 0.2676\n"
     ]
    }
   ],
   "source": [
    "loss, mse, mae = tf_model.evaluate(X_test, y_test)\n",
    "prediction = tf_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "208e7c75",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_evaluation(tf_model, X_test, y_test, \"TensorFlow\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f94a0da",
   "metadata": {},
   "source": [
    "## Plot some example predictions\n",
    "\n",
    "Let's plot the molecular structure and meaasured/predicted logP values. We'll use the same indicies as we did for the Ridge \n",
    "Regression."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "b3f7f1a5",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CN(C)C(N)N\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224831759391296\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224831759391296\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224831759391296 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224831759391296\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224831759391296 = $3Dmol.createViewer($(\"#3dmolviewer_16224831759391296\"),{backgroundColor:\"white\"});\nviewer_16224831759391296.zoomTo();\n\tviewer_16224831759391296.addModel(\"\\n     RDKit          2D\\n\\n 17 16  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500   -1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.5490    0.5490    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.9510   -2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0653   -0.9530    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.9510    2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.8481    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.6462   -0.2422    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  2  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  3 10  1  0\\n  3 11  1  0\\n  3 12  1  0\\n  4 13  1  0\\n  5 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\n  6 17  1  0\\nM  END\\n\",\"mol\");\n\tviewer_16224831759391296.setStyle({\"stick\": {}});\n\tviewer_16224831759391296.rotate(45,\"y\");\n\tviewer_16224831759391296.zoomTo();\nviewer_16224831759391296.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224831759391296\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224831759391296\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224831759391296 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224831759391296\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224831759391296 = $3Dmol.createViewer($(\"#3dmolviewer_16224831759391296\"),{backgroundColor:\"white\"});\n",
       "viewer_16224831759391296.zoomTo();\n",
       "\tviewer_16224831759391296.addModel(\"\\n     RDKit          2D\\n\\n 17 16  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500   -1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.5490    0.5490    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.9510   -2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0653   -0.9530    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.9510    2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.8481    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.6462   -0.2422    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  2  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  3 10  1  0\\n  3 11  1  0\\n  3 12  1  0\\n  4 13  1  0\\n  5 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\n  6 17  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224831759391296.setStyle({\"stick\": {}});\n",
       "\tviewer_16224831759391296.rotate(45,\"y\");\n",
       "\tviewer_16224831759391296.zoomTo();\n",
       "viewer_16224831759391296.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: C1=NN=[C+]S1\n",
      "Target logP = 0.400, Prediction logP = 0.233\n",
      "==================================================\n",
      "SMILES notation: C(C(O)ON)O\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224831759672282\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224831759672282\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224831759672282 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224831759672282\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224831759672282 = $3Dmol.createViewer($(\"#3dmolviewer_16224831759672282\"),{backgroundColor:\"white\"});\nviewer_16224831759672282.zoomTo();\n\tviewer_16224831759672282.addModel(\"\\n     RDKit          2D\\n\\n 13 12  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.7990   -2.2500    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.7990   -3.7500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.0981   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -2.2500   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  2  4  1  0\\n  4  5  1  0\\n  1  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  2  9  1  0\\n  3 10  1  0\\n  5 11  1  0\\n  5 12  1  0\\n  6 13  1  0\\nM  END\\n\",\"mol\");\n\tviewer_16224831759672282.setStyle({\"stick\": {}});\n\tviewer_16224831759672282.rotate(45,\"y\");\n\tviewer_16224831759672282.zoomTo();\nviewer_16224831759672282.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224831759672282\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224831759672282\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224831759672282 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224831759672282\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224831759672282 = $3Dmol.createViewer($(\"#3dmolviewer_16224831759672282\"),{backgroundColor:\"white\"});\n",
       "viewer_16224831759672282.zoomTo();\n",
       "\tviewer_16224831759672282.addModel(\"\\n     RDKit          2D\\n\\n 13 12  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.7990   -2.2500    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.7990   -3.7500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.0981   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -2.2500   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  2  4  1  0\\n  4  5  1  0\\n  1  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  2  9  1  0\\n  3 10  1  0\\n  5 11  1  0\\n  5 12  1  0\\n  6 13  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224831759672282.setStyle({\"stick\": {}});\n",
       "\tviewer_16224831759672282.rotate(45,\"y\");\n",
       "\tviewer_16224831759672282.zoomTo();\n",
       "viewer_16224831759672282.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: COCCCNN\n",
      "Target logP = -0.700, Prediction logP = -0.729\n",
      "==================================================\n",
      "SMILES notation: C(=O)(N)I\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224831759988039\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224831759988039\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224831759988039 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224831759988039\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224831759988039 = $3Dmol.createViewer($(\"#3dmolviewer_16224831759988039\"),{backgroundColor:\"white\"});\nviewer_16224831759988039.zoomTo();\n\tviewer_16224831759988039.addModel(\"\\n     RDKit          2D\\n\\n  6  5  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 I   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.5981   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  2  0\\n  1  3  1  0\\n  1  4  1  0\\n  3  5  1  0\\n  3  6  1  0\\nM  END\\n\",\"mol\");\n\tviewer_16224831759988039.setStyle({\"stick\": {}});\n\tviewer_16224831759988039.rotate(45,\"y\");\n\tviewer_16224831759988039.zoomTo();\nviewer_16224831759988039.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224831759988039\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224831759988039\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224831759988039 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224831759988039\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224831759988039 = $3Dmol.createViewer($(\"#3dmolviewer_16224831759988039\"),{backgroundColor:\"white\"});\n",
       "viewer_16224831759988039.zoomTo();\n",
       "\tviewer_16224831759988039.addModel(\"\\n     RDKit          2D\\n\\n  6  5  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 I   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.5981   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  2  0\\n  1  3  1  0\\n  1  4  1  0\\n  3  5  1  0\\n  3  6  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224831759988039.setStyle({\"stick\": {}});\n",
       "\tviewer_16224831759988039.rotate(45,\"y\");\n",
       "\tviewer_16224831759988039.zoomTo();\n",
       "viewer_16224831759988039.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CS(=O)P\n",
      "Target logP = -0.600, Prediction logP = -0.099\n",
      "==================================================\n",
      "SMILES notation: CCCCOCC\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224831760339534\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224831760339534\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224831760339534 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224831760339534\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224831760339534 = $3Dmol.createViewer($(\"#3dmolviewer_16224831760339534\"),{backgroundColor:\"white\"});\nviewer_16224831760339534.zoomTo();\n\tviewer_16224831760339534.addModel(\"\\n     RDKit          2D\\n\\n 21 20  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.7500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    2.5981    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.4510    2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.0490    0.5490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.2010    3.3481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.7990    1.8481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  3 14  1  0\\n  4 15  1  0\\n  4 16  1  0\\n  6 17  1  0\\n  6 18  1  0\\n  7 19  1  0\\n  7 20  1  0\\n  7 21  1  0\\nM  END\\n\",\"mol\");\n\tviewer_16224831760339534.setStyle({\"stick\": {}});\n\tviewer_16224831760339534.rotate(45,\"y\");\n\tviewer_16224831760339534.zoomTo();\nviewer_16224831760339534.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224831760339534\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224831760339534\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224831760339534 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224831760339534\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224831760339534 = $3Dmol.createViewer($(\"#3dmolviewer_16224831760339534\"),{backgroundColor:\"white\"});\n",
       "viewer_16224831760339534.zoomTo();\n",
       "\tviewer_16224831760339534.addModel(\"\\n     RDKit          2D\\n\\n 21 20  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.7500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    2.5981    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.4510    2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.0490    0.5490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.2010    3.3481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    8.7990    1.8481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  3 14  1  0\\n  4 15  1  0\\n  4 16  1  0\\n  6 17  1  0\\n  6 18  1  0\\n  7 19  1  0\\n  7 20  1  0\\n  7 21  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224831760339534.setStyle({\"stick\": {}});\n",
       "\tviewer_16224831760339534.rotate(45,\"y\");\n",
       "\tviewer_16224831760339534.zoomTo();\n",
       "viewer_16224831760339534.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CP1[SH2]C[SH2]1\n",
      "Target logP = 0.700, Prediction logP = 0.707\n",
      "==================================================\n",
      "SMILES notation: [C](F)(F)Br\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224831760621583\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224831760621583\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224831760621583 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224831760621583\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224831760621583 = $3Dmol.createViewer($(\"#3dmolviewer_16224831760621583\"),{backgroundColor:\"white\"});\nviewer_16224831760621583.zoomTo();\n\tviewer_16224831760621583.addModel(\"\\n     RDKit          2D\\n\\n  4  3  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  3  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 F   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 F   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 Br  0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  1  3  1  0\\n  1  4  1  0\\nM  RAD  1   1   2\\nM  END\\n\",\"mol\");\n\tviewer_16224831760621583.setStyle({\"stick\": {}});\n\tviewer_16224831760621583.rotate(45,\"y\");\n\tviewer_16224831760621583.zoomTo();\nviewer_16224831760621583.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224831760621583\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224831760621583\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224831760621583 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224831760621583\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224831760621583 = $3Dmol.createViewer($(\"#3dmolviewer_16224831760621583\"),{backgroundColor:\"white\"});\n",
       "viewer_16224831760621583.zoomTo();\n",
       "\tviewer_16224831760621583.addModel(\"\\n     RDKit          2D\\n\\n  4  3  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  3  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 F   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 F   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 Br  0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  1  3  1  0\\n  1  4  1  0\\nM  RAD  1   1   2\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224831760621583.setStyle({\"stick\": {}});\n",
       "\tviewer_16224831760621583.rotate(45,\"y\");\n",
       "\tviewer_16224831760621583.zoomTo();\n",
       "viewer_16224831760621583.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CNSSNC\n",
      "Target logP = 0.400, Prediction logP = 0.311\n",
      "==================================================\n",
      "SMILES notation: C(C[NH3+])C(O)O\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_1622483176089943\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_1622483176089943\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_1622483176089943 = null;\nvar warn = document.getElementById(\"3dmolwarning_1622483176089943\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_1622483176089943 = $3Dmol.createViewer($(\"#3dmolviewer_1622483176089943\"),{backgroundColor:\"white\"});\nviewer_1622483176089943.zoomTo();\n\tviewer_1622483176089943.addModel(\"\\n     RDKit          2D\\n\\n 16 15  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 N   0  0  0  0  0  4  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -3.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000   -1.5000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -3.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -2.7990   -2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  1  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  2  9  1  0\\n  2 10  1  0\\n  3 11  1  0\\n  3 12  1  0\\n  3 13  1  0\\n  4 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\nM  CHG  1   3   1\\nM  END\\n\",\"mol\");\n\tviewer_1622483176089943.setStyle({\"stick\": {}});\n\tviewer_1622483176089943.rotate(45,\"y\");\n\tviewer_1622483176089943.zoomTo();\nviewer_1622483176089943.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_1622483176089943\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_1622483176089943\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_1622483176089943 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_1622483176089943\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_1622483176089943 = $3Dmol.createViewer($(\"#3dmolviewer_1622483176089943\"),{backgroundColor:\"white\"});\n",
       "viewer_1622483176089943.zoomTo();\n",
       "\tviewer_1622483176089943.addModel(\"\\n     RDKit          2D\\n\\n 16 15  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 N   0  0  0  0  0  4  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -3.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000   -1.5000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -3.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -2.7990   -2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  1  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  2  9  1  0\\n  2 10  1  0\\n  3 11  1  0\\n  3 12  1  0\\n  3 13  1  0\\n  4 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\nM  CHG  1   3   1\\nM  END\\n\",\"mol\");\n",
       "\tviewer_1622483176089943.setStyle({\"stick\": {}});\n",
       "\tviewer_1622483176089943.rotate(45,\"y\");\n",
       "\tviewer_1622483176089943.zoomTo();\n",
       "viewer_1622483176089943.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CC([CH2])(Cl)Cl\n",
      "Target logP = 2.100, Prediction logP = 1.967\n",
      "==================================================\n",
      "SMILES notation: C=CCCCl\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224831761177025\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224831761177025\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224831761177025 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224831761177025\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224831761177025 = $3Dmol.createViewer($(\"#3dmolviewer_16224831761177025\"),{backgroundColor:\"white\"});\nviewer_16224831761177025.zoomTo();\n\tviewer_16224831761177025.addModel(\"\\n     RDKit          2D\\n\\n 12 11  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.5981   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.8971   -0.7500    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.1962   -1.5000    0.0000 Cl  0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.3481    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.8481   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.1471   -2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.6471    0.5490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  2  3\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  1  6  1  0\\n  1  7  1  0\\n  2  8  1  0\\n  3  9  1  0\\n  3 10  1  0\\n  4 11  1  0\\n  4 12  1  0\\nM  END\\n\",\"mol\");\n\tviewer_16224831761177025.setStyle({\"stick\": {}});\n\tviewer_16224831761177025.rotate(45,\"y\");\n\tviewer_16224831761177025.zoomTo();\nviewer_16224831761177025.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224831761177025\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224831761177025\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224831761177025 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224831761177025\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224831761177025 = $3Dmol.createViewer($(\"#3dmolviewer_16224831761177025\"),{backgroundColor:\"white\"});\n",
       "viewer_16224831761177025.zoomTo();\n",
       "\tviewer_16224831761177025.addModel(\"\\n     RDKit          2D\\n\\n 12 11  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    0.7500    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.5981   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.8971   -0.7500    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.1962   -1.5000    0.0000 Cl  0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.2990    0.7500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.2990    2.2500    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.3481    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.8481   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.1471   -2.0490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.6471    0.5490    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  2  3\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  1  6  1  0\\n  1  7  1  0\\n  2  8  1  0\\n  3  9  1  0\\n  3 10  1  0\\n  4 11  1  0\\n  4 12  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224831761177025.setStyle({\"stick\": {}});\n",
       "\tviewer_16224831761177025.rotate(45,\"y\");\n",
       "\tviewer_16224831761177025.zoomTo();\n",
       "viewer_16224831761177025.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: [2H]C([2H])([2H])C([2H])(CO)C([2H])([2H])[2H]\n",
      "Target logP = 0.800, Prediction logP = 0.926\n",
      "==================================================\n",
      "SMILES notation: CCO[S+](C)C\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224831761461015\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224831761461015\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224831761461015 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224831761461015\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224831761461015 = $3Dmol.createViewer($(\"#3dmolviewer_16224831761461015\"),{backgroundColor:\"white\"});\nviewer_16224831761461015.zoomTo();\n\tviewer_16224831761461015.addModel(\"\\n     RDKit          2D\\n\\n 17 16  0  0  0  0  0  0  0  0999 V2000\\n    0.0000   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 S   0  0  0  0  0  3  0  0  0  0  0  0\\n    5.2500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.3519    1.2059    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    2.7990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500   -0.2010    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.8816    2.1234    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.2990    3.3481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  2 10  1  0\\n  2 11  1  0\\n  5 12  1  0\\n  5 13  1  0\\n  5 14  1  0\\n  6 15  1  0\\n  6 16  1  0\\n  6 17  1  0\\nM  CHG  1   4   1\\nM  END\\n\",\"mol\");\n\tviewer_16224831761461015.setStyle({\"stick\": {}});\n\tviewer_16224831761461015.rotate(45,\"y\");\n\tviewer_16224831761461015.zoomTo();\nviewer_16224831761461015.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224831761461015\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224831761461015\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224831761461015 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224831761461015\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224831761461015 = $3Dmol.createViewer($(\"#3dmolviewer_16224831761461015\"),{backgroundColor:\"white\"});\n",
       "viewer_16224831761461015.zoomTo();\n",
       "\tviewer_16224831761461015.addModel(\"\\n     RDKit          2D\\n\\n 17 16  0  0  0  0  0  0  0  0999 V2000\\n    0.0000   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 S   0  0  0  0  0  3  0  0  0  0  0  0\\n    5.2500    1.2990    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.3519    1.2059    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    2.7990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500   -0.2010    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    2.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.8816    2.1234    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.2990    3.3481    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  4  6  1  0\\n  1  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  2 10  1  0\\n  2 11  1  0\\n  5 12  1  0\\n  5 13  1  0\\n  5 14  1  0\\n  6 15  1  0\\n  6 16  1  0\\n  6 17  1  0\\nM  CHG  1   4   1\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224831761461015.setStyle({\"stick\": {}});\n",
       "\tviewer_16224831761461015.rotate(45,\"y\");\n",
       "\tviewer_16224831761461015.zoomTo();\n",
       "viewer_16224831761461015.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: CNC(CS)N\n",
      "Target logP = -0.700, Prediction logP = -0.293\n",
      "==================================================\n",
      "SMILES notation: [13CH3][13CH2][13CH2][13CH2][13CH2][13CH2]O\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_16224831761738298\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_16224831761738298\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_16224831761738298 = null;\nvar warn = document.getElementById(\"3dmolwarning_16224831761738298\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_16224831761738298 = $3Dmol.createViewer($(\"#3dmolviewer_16224831761738298\"),{backgroundColor:\"white\"});\nviewer_16224831761738298.zoomTo();\n\tviewer_16224831761738298.addModel(\"\\n     RDKit          2D\\n\\n 21 20  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    9.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    9.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  3 14  1  0\\n  4 15  1  0\\n  4 16  1  0\\n  5 17  1  0\\n  5 18  1  0\\n  6 19  1  0\\n  6 20  1  0\\n  7 21  1  0\\nM  ISO  6   1  13   2  13   3  13   4  13   5  13   6  13\\nM  END\\n\",\"mol\");\n\tviewer_16224831761738298.setStyle({\"stick\": {}});\n\tviewer_16224831761738298.rotate(45,\"y\");\n\tviewer_16224831761738298.zoomTo();\nviewer_16224831761738298.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_16224831761738298\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_16224831761738298\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_16224831761738298 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_16224831761738298\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_16224831761738298 = $3Dmol.createViewer($(\"#3dmolviewer_16224831761738298\"),{backgroundColor:\"white\"});\n",
       "viewer_16224831761738298.zoomTo();\n",
       "\tviewer_16224831761738298.addModel(\"\\n     RDKit          2D\\n\\n 21 20  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    9.0000    0.0000    0.0000 O   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    4.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    9.7500    1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  3 14  1  0\\n  4 15  1  0\\n  4 16  1  0\\n  5 17  1  0\\n  5 18  1  0\\n  6 19  1  0\\n  6 20  1  0\\n  7 21  1  0\\nM  ISO  6   1  13   2  13   3  13   4  13   5  13   6  13\\nM  END\\n\",\"mol\");\n",
       "\tviewer_16224831761738298.setStyle({\"stick\": {}});\n",
       "\tviewer_16224831761738298.rotate(45,\"y\");\n",
       "\tviewer_16224831761738298.zoomTo();\n",
       "viewer_16224831761738298.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: COCC(O)I\n",
      "Target logP = 0.200, Prediction logP = 0.383\n",
      "==================================================\n",
      "SMILES notation: CCNNNNI\n",
      "Rotate me!\n"
     ]
    },
    {
     "data": {
      "application/3dmoljs_load.v0": "<div id=\"3dmolviewer_1622483176204109\"  style=\"position: relative; width: 300px; height: 300px\">\n        <p id=\"3dmolwarning_1622483176204109\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n        </div>\n<script>\n\nvar loadScriptAsync = function(uri){\n  return new Promise((resolve, reject) => {\n    var tag = document.createElement('script');\n    tag.src = uri;\n    tag.async = true;\n    tag.onload = () => {\n      resolve();\n    };\n  var firstScriptTag = document.getElementsByTagName('script')[0];\n  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n});\n};\n\nif(typeof $3Dmolpromise === 'undefined') {\n$3Dmolpromise = null;\n  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n}\n\nvar viewer_1622483176204109 = null;\nvar warn = document.getElementById(\"3dmolwarning_1622483176204109\");\nif(warn) {\n    warn.parentNode.removeChild(warn);\n}\n$3Dmolpromise.then(function() {\nviewer_1622483176204109 = $3Dmol.createViewer($(\"#3dmolviewer_1622483176204109\"),{backgroundColor:\"white\"});\nviewer_1622483176204109.zoomTo();\n\tviewer_1622483176204109.addModel(\"\\n     RDKit          2D\\n\\n 16 15  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    1.2990    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    2.5981    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    2.5981    0.0000 I   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  4 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\nM  END\\n\",\"mol\");\n\tviewer_1622483176204109.setStyle({\"stick\": {}});\n\tviewer_1622483176204109.rotate(45,\"y\");\n\tviewer_1622483176204109.zoomTo();\nviewer_1622483176204109.render();\n});\n</script>",
      "text/html": [
       "<div id=\"3dmolviewer_1622483176204109\"  style=\"position: relative; width: 300px; height: 300px\">\n",
       "        <p id=\"3dmolwarning_1622483176204109\" style=\"background-color:#ffcccc;color:black\">You appear to be running in JupyterLab (or JavaScript failed to load for some other reason).  You need to install the 3dmol extension: <br>\n",
       "        <tt>jupyter labextension install jupyterlab_3dmol</tt></p>\n",
       "        </div>\n",
       "<script>\n",
       "\n",
       "var loadScriptAsync = function(uri){\n",
       "  return new Promise((resolve, reject) => {\n",
       "    var tag = document.createElement('script');\n",
       "    tag.src = uri;\n",
       "    tag.async = true;\n",
       "    tag.onload = () => {\n",
       "      resolve();\n",
       "    };\n",
       "  var firstScriptTag = document.getElementsByTagName('script')[0];\n",
       "  firstScriptTag.parentNode.insertBefore(tag, firstScriptTag);\n",
       "});\n",
       "};\n",
       "\n",
       "if(typeof $3Dmolpromise === 'undefined') {\n",
       "$3Dmolpromise = null;\n",
       "  $3Dmolpromise = loadScriptAsync('https://3dmol.org/build/3Dmol.js');\n",
       "}\n",
       "\n",
       "var viewer_1622483176204109 = null;\n",
       "var warn = document.getElementById(\"3dmolwarning_1622483176204109\");\n",
       "if(warn) {\n",
       "    warn.parentNode.removeChild(warn);\n",
       "}\n",
       "$3Dmolpromise.then(function() {\n",
       "viewer_1622483176204109 = $3Dmol.createViewer($(\"#3dmolviewer_1622483176204109\"),{backgroundColor:\"white\"});\n",
       "viewer_1622483176204109.zoomTo();\n",
       "\tviewer_1622483176204109.addModel(\"\\n     RDKit          2D\\n\\n 16 15  0  0  0  0  0  0  0  0999 V2000\\n    0.0000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    0.0000    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    0.0000    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500    1.2990    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    1.2990    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000    2.5981    0.0000 N   0  0  0  0  0  0  0  0  0  0  0  0\\n    7.5000    2.5981    0.0000 I   0  0  0  0  0  0  0  0  0  0  0  0\\n   -1.5000    0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    0.0000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n   -0.0000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000   -1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    1.5000    1.5000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.7500   -1.2990    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    3.0000    2.5981    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    6.0000   -0.0000    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n    5.2500    3.8971    0.0000 H   0  0  0  0  0  0  0  0  0  0  0  0\\n  1  2  1  0\\n  2  3  1  0\\n  3  4  1  0\\n  4  5  1  0\\n  5  6  1  0\\n  6  7  1  0\\n  1  8  1  0\\n  1  9  1  0\\n  1 10  1  0\\n  2 11  1  0\\n  2 12  1  0\\n  3 13  1  0\\n  4 14  1  0\\n  5 15  1  0\\n  6 16  1  0\\nM  END\\n\",\"mol\");\n",
       "\tviewer_1622483176204109.setStyle({\"stick\": {}});\n",
       "\tviewer_1622483176204109.rotate(45,\"y\");\n",
       "\tviewer_1622483176204109.zoomTo();\n",
       "viewer_1622483176204109.render();\n",
       "});\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SMILES notation: [CH2]CC(CO)N\n",
      "Target logP = -0.400, Prediction logP = -0.607\n",
      "==================================================\n"
     ]
    }
   ],
   "source": [
    "for idx in indicies:\n",
    "    plot_example(tf_model, X_test, y_test, df, idx, show_3d=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2bd23f7",
   "metadata": {},
   "source": [
    "## Let's finally compare the TensorFlow model to the Ridge Regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "d7bd97b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_both_evaluation(tf_model, ridge_model, X_test, y_test):\n",
    "    \"\"\"\n",
    "    Scatter plot between target and prediction logP values for both models\n",
    "    \"\"\"\n",
    "    \n",
    "    prediction_tf = tf_model.predict(X_test)\n",
    "    mae_tf = mean_absolute_error(y_test, prediction_tf)\n",
    "    mse_tf = mean_squared_error(y_test, prediction_tf)\n",
    "    \n",
    "    prediction_ridge = ridge_model.predict(X_test)\n",
    "    mae_ridge = mean_absolute_error(y_test, prediction_ridge)\n",
    "    mse_ridge = mean_squared_error(y_test, prediction_ridge)\n",
    "\n",
    "    plt.figure(figsize=(20,10))\n",
    "    plt.subplot(1,2,2);\n",
    "    sns.regplot(x=prediction_tf, y=y_test, scatter_kws={\"color\": \"blue\"}, line_kws={\"color\": \"red\"});\n",
    "    plt.xlabel('Model Prediction', fontsize=24);\n",
    "    plt.ylabel('Measured logP', fontsize=24);\n",
    "    plt.xticks(fontsize=24);\n",
    "    plt.yticks(fontsize=24);\n",
    "    plt.suptitle('Predicting logP from SMILES', fontsize=32);\n",
    "    plt.title(f'TensorFlow: MAE = {mae_tf:.3f}, MSE = {mse_tf:.3f}', fontsize=24);\n",
    "    plt.grid(True);\n",
    "    cc = np.corrcoef(y_test, np.squeeze(prediction_tf))\n",
    "    plt.text(-1.5, 2, f\"Correlation\\ncoefficient\\n{cc[0][1]:.4f}\", fontsize=24, \n",
    "             horizontalalignment=\"center\", backgroundcolor=\"black\", color=\"white\")\n",
    "    \n",
    "    plt.subplot(1,2,1);\n",
    "    sns.regplot(x=prediction_ridge, y=y_test, scatter_kws={\"color\": \"blue\"}, line_kws={\"color\": \"red\"});\n",
    "    plt.xlabel('Model Prediction', fontsize=24);\n",
    "    plt.ylabel('Measured logP', fontsize=24);\n",
    "    plt.xticks(fontsize=24);\n",
    "    plt.yticks(fontsize=24);\n",
    "    plt.suptitle('Predicting logP from SMILES', fontsize=32);\n",
    "    plt.title(f'Ridge MAE = {mae_ridge:.3f}, MSE = {mse_ridge:.3f}', fontsize=24);\n",
    "    \n",
    "    cc = np.corrcoef(y_test, np.squeeze(prediction_ridge))\n",
    "    plt.text(-1.5, 2, f\"Correlation\\ncoefficient\\n{cc[0][1]:.4f}\", fontsize=24, \n",
    "             horizontalalignment=\"center\", backgroundcolor=\"black\", color=\"white\")\n",
    "    plt.grid(True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "3dbb85ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1440x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_both_evaluation(tf_model, ridge_model, X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9a44b87",
   "metadata": {},
   "source": [
    "## OpenVINO™-TensorFlow Bridge\n",
    "\n",
    "The [OpenVINO™-TensorFlow](https://github.com/openvinotoolkit/openvino_tensorflow) bridge optimizes deep learning inference on Intel® hardware. It is designed for developers who want to get started using OpenVINO™ with minimal modifications to their inference code. OpenVINO™ integration with TensorFlow accelerates inference across many AI models on a variety of Intel® silicon such as:\n",
    "\n",
    "+ Intel® CPUs\n",
    "+ Intel® integrated GPUs\n",
    "+ Intel® Movidius™ Vision Processing Units - referred as VPU\n",
    "+ Intel® Vision Accelerator Design with 8 Intel Movidius™ MyriadX VPUs - referred as VAD-M or HDDL\n",
    "\n",
    "The current TensorFlow model is a trivial multi-layered network, but for more complex TensorFlow topologies there may be a significant speedup of inference. I'd encourage you to try other deep learning topologies to evaluate the speedups. Note: The truly optimal speedup on Intel® hardware can be obtained by converting the TensorFlow model to the [OpenVINO™](https://docs.openvinotoolkit.org/latest/index.html) format and using the native OpenVINO inference engine.\n",
    "\n",
    "To install the OpenVINO™-TensorFlow bridge you can find specific directions [here](https://openvinotoolkit.github.io/openvino_tensorflow/)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "1ca72a8b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TensorFlow version:  2.4.1\n",
      "OpenVINO integration with TensorFlow version: b'0.5.0'\n",
      "nGraph version used for this build: b'0.0.0+b33800a'\n",
      "TensorFlow version used for this build: v2.4.0-49-g85c8b2a817f\n",
      "CXX11_ABI flag used for this build: 0\n",
      "OpenVINO integration with TensorFlow built with Grappler: False\n",
      "\n",
      "OpenVINO is enabled: True\n"
     ]
    }
   ],
   "source": [
    "#!pip install openvino-tensorflow==0.5.0\n",
    "print('TensorFlow version: ', tf.__version__)\n",
    "import openvino_tensorflow\n",
    "print(openvino_tensorflow.__version__)\n",
    "\n",
    "print(f\"OpenVINO is enabled: {openvino_tensorflow.is_enabled()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3d5da69",
   "metadata": {},
   "source": [
    "## Chosing the AI Hardware\n",
    "The next command prints all available hardware for AI inference using OpenVINO™. We can easily switch between hardware using the `set_backend` command. That's it. The bridge does the rest."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "a86dd1bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['CPU']"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "openvino_tensorflow.list_backends()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "ff95d753",
   "metadata": {},
   "outputs": [],
   "source": [
    "openvino_tensorflow.set_backend('CPU')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "c3c6ebd6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "46/46 [==============================] - 0s 656us/step - loss: 0.1244 - mse: 0.1244 - mae: 0.2676\n"
     ]
    }
   ],
   "source": [
    "loss, mse, mae = tf_model.evaluate(X_test, y_test)\n",
    "prediction_ov = tf_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "2a0129ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The maximum difference between OpenVINO and TensorFlow predictions is 0.00000\n"
     ]
    }
   ],
   "source": [
    "print(f\"The maximum difference between OpenVINO and TensorFlow predictions is {np.max(prediction_ov - prediction):.5f}\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}