IntroQueries.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "from citrination_client import CitrinationClient\n",
    "from os import environ\n",
    "\n",
    "client = CitrinationClient(environ['CITRINATION_API_KEY'], 'https://citrination.com')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Introduction to Queries\n",
    "\n",
    "Queries allow you to search and download PIFs from Citrination.  By basing scripts on queries intead of local files, you can:\n",
    " 1. Share data with your group without running internal infrastructure or worrying about portability\n",
    " 1. Expand your queries to include data uploaded by other groups in your field\n",
    " 1. Publish post-processing scripts backed by Citrination data to improve reproducibility"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Query interfaces\n",
    "\n",
    "There are multiple interfaces to a unified backend based on elastic search:\n",
    " 1. The website\n",
    " 1. The python and java citrination clients \n",
    " 1. A REST API \n",
    " \n",
    "This tutorial will focus on the python citrination client"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## `PifSystemReturningQuery`\n",
    "\n",
    "Each search and download request is stored in a query object: the `PifSystemReturningQuery`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A PifQuery is structured in the following hierarchical way:\n",
    "\n",
    "```\n",
    "PifSystemReturningQuery(\n",
    "            size=0,\n",
    "            query=DataQuery(\n",
    "                dataset=DatasetQuery(\n",
    "                    id=Filter(equal='151278')\n",
    "                ),\n",
    "                system=PifSystemQuery(\n",
    "                    chemical_formula=ChemicalFieldQuery(\n",
    "                        filter=ChemicalFilter(\n",
    "                            equal='C22H15NSSi')))))\n",
    "                            ```\n",
    "                            \n",
    "see our API docs for more information: https://citrineinformatics.github.io/api-documentation/#tag/search"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you are working through the notebooks in the order they are presented on our [API Examples](https://citrination.org/learn/api-examples/) site, use the `dataset_id` from the previous Importing VASP calculations notebook in the query below. \n",
    "\n",
    "Otherwise, we can start with the [Bandgaps Dataset](https://citrination.com/datasets/1160/show_search), but feel free to use any public dataset.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 2 PIFs in dataset 154987\n"
     ]
    }
   ],
   "source": [
    "from citrination_client import PifSystemReturningQuery, PifSystemQuery\n",
    "from citrination_client import DataQuery, DatasetQuery, ChemicalFieldQuery, ChemicalFilter, Filter\n",
    "from citrination_client import PropertyQuery, FieldQuery\n",
    "\n",
    "#change dataset id below\n",
    "my_dataset_id = 154987\n",
    "bandgaps_dataset_id = 1160\n",
    "\n",
    "query_dataset = PifSystemReturningQuery(size=5, \n",
    "                                        query=DataQuery(\n",
    "                                            dataset=DatasetQuery(\n",
    "                                                id=Filter(equal=str(my_dataset_id))\n",
    "                                         )))\n",
    "query_result = client.search.pif_search(query_dataset)\n",
    "print(\"Found {} PIFs in dataset {}\".format(query_result.total_num_hits, my_dataset_id))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "The result contains the matching PIFs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{\n",
      "  \"tags\": [\n",
      "    \"my_first_upload\"\n",
      "  ],\n",
      "  \"category\": \"system.chemical\",\n",
      "  \"qualityReport\": {\n",
      "    \"body\": \"\",\n",
      "    \"name\": \"\",\n",
      "    \"postamble\": \"nowrap Summary:       Number of tests:   6\\n                critical error:   0\\n                         error:   0\\n               serious warning:   0\\n                       warning:   2\\n                       caution:   0\\n                recommendation:   1\\n                   information:   1\\n                       comment:   0\\n                            ok:   2\\n                not applicab\n"
     ]
    }
   ],
   "source": [
    "from pypif.pif import dumps\n",
    "pifs = [x.system for x in query_result.hits]\n",
    "print(dumps(pifs[0], indent=2)[:550])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "A `simple_chemical_search` query using the citrination client can accept lists of datasets to bring in data from collaborators:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 2 PIFs in datasets 154987 and 1160\n"
     ]
    }
   ],
   "source": [
    "my_dataset_id = 154987; another_dataset_id = 1160\n",
    "\n",
    "query = client.generate_simple_chemical_query(include_datasets=[my_dataset_id, another_dataset_id], size=500)\n",
    "retuls = client.search.pif_search(query)\n",
    "print(\"Found {} PIFs in datasets {} and {}\".format(\n",
    "        query_result.total_num_hits, my_dataset_id, another_dataset_id\n",
    "    ))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "nbpresent": {
     "id": "f2f733dc-95c8-4751-a337-7bba7fcbe4d7"
    },
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Advanced Queries\n",
    "\n",
    "Writing analysis and post-processing on top of PIFs lets us:\n",
    " 1. Pull in data from collaborators and published sources\n",
    " 1. Share our methodology with other researchers with data in PIFs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "Let's actually do that!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Real example: phase stability diagram\n",
    "\n",
    "There are surely more Al-Cu binaries on Citrination.  We will:\n",
    " 1. Find relevant data via the web interface\n",
    " 2. Filter the data programatically with the query\n",
    " 3. Extract the data using query extractions\n",
    " 4. Add the data to our plot from before"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "nbpresent": {
     "id": "c41a4f53-9a6a-435e-b554-90780aa20fcd"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Search for relevant data\n",
    "\n",
    "The exploratory search is usually best done on the website.  Tips and tricks:\n",
    " * \"AlxCuy\" searches for any Al-Cu binary\n",
    " * \"energy\" searches for properties with \"energy\" in their name"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Adding filters to the query\n",
    "\n",
    "Now that we are querying beyond just our data, we need to filter down to records that contain Al-Cu energies.\n",
    "\n",
    "### Filter by chemical formula\n",
    "\n",
    "1. You can filter over known datasets\n",
    "2. You can filter over all public data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n"
     ]
    }
   ],
   "source": [
    "#filtering over a known dataset by formula='CdTe'\n",
    "\n",
    "res = client.pif_search(PifSystemReturningQuery(\n",
    "            size=0,\n",
    "            query=DataQuery(\n",
    "                dataset=DatasetQuery(\n",
    "                    id=Filter(equal='1160')\n",
    "                ),\n",
    "                system=PifSystemQuery(\n",
    "                    chemical_formula=ChemicalFieldQuery(\n",
    "                        filter=ChemicalFilter(\n",
    "                            equal='AlxCuy'))))))\n",
    "print(res.total_num_hits)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 139 PIFs\n"
     ]
    }
   ],
   "source": [
    "#filtering by formula='AlxCuy'\n",
    "\n",
    "query = PifSystemReturningQuery(\n",
    "            size=10,\n",
    "            query=DataQuery(\n",
    "                simple='AlxCuy'))\n",
    "\n",
    "query_result = client.search.pif_search(query)\n",
    "print(\"Found {} PIFs\".format(query_result.total_num_hits))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "### Filter by property\n",
    "\n",
    "We can do quick filtered searches with `citrination_client.simple_chemical_search`. It returns a `PifSearchResult` object as before, but with useful filter flags. See the docs for more info.\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 5 PIFs\n",
      "Found 16 PIFs\n"
     ]
    }
   ],
   "source": [
    "query = client.generate_simple_chemical_query(property_name='Formation Energy', chemical_formula='AlxCuy', size=500)\n",
    "quicksearch1 = client.search.pif_search(query)\n",
    "print(\"Found {} PIFs\".format(quicksearch1.total_num_hits))\n",
    "\n",
    "query = client.generate_simple_chemical_query(property_name=['Formation Energy', 'Enthalpy of Formation'], chemical_formula='AlxCuy', size=500)\n",
    "quicksearch2 = client.search.pif_search(query)\n",
    "print(\"Found {} PIFs\".format(quicksearch2.total_num_hits))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "nbpresent": {
     "id": "45a1cd9a-989f-44e6-8753-79a4abcfbe28"
    },
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{\n",
      "  \"category\": \"system.chemical\",\n",
      "  \"references\": [\n",
      "    {\n",
      "      \"doi\": \"10.1016/j.commatsci.2012.02.002\"\n",
      "    },\n",
      "    {\n",
      "      \"doi\": \"10.1016/j.commatsci.2012.02.005\"\n",
      "    }\n",
      "  ],\n",
      "  \"uid\": \"5CA6B3488E258DB2E0A73ADDA2F97C4D\",\n",
      "  \"properties\": [\n",
      "    {\n",
      "      \"name\": \"Structure designation\",\n",
      "      \"scalars\": [\n",
      "        {\n",
      "          \"value\": \"CaF[2]\"\n",
      "        }\n",
      "      ]\n",
      "    },\n",
      "    {\n",
      "      \"name\": \"Formation energy\",\n",
      "      \"scalars\": [\n",
      "        {\n",
      "          \"value\": \"-0.1799\"\n",
      "        }\n",
      "      ],\n",
      "      \"units\": \"eV/atom\",\n",
      "      \"dataType\": \"COMPUTATIONAL\"\n",
      "    }\n",
      "  ],\n",
      "  \"chemicalFormula\": \"CuAl2\"\n",
      "}\n"
     ]
    }
   ],
   "source": [
    "from pypif.pif import dumps\n",
    "print(dumps(quicksearch1.hits[1].system, indent=2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Extract data from results using the query language\n",
    "\n",
    "Multiple sources typically use multiple internal formats.  The query language is smart about crawling the PIFs for properties with the right names, while `pypif` is not.  Fortunately, you can use the query language to do data extraction as well!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "nbpresent": {
     "id": "261197d6-24ad-4083-8df9-33ba39ff56b7"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "### Using `extract_as`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 16 PIFs\n",
      "[{'formation_enthalpy': '-0.1889', 'formula': 'Cu3Al'}, {'formation_enthalpy': '-0.1799', 'formula': 'CuAl2'}]\n"
     ]
    }
   ],
   "source": [
    "query = PifSystemReturningQuery(\n",
    "            size=500,\n",
    "            query=DataQuery(\n",
    "                system=PifSystemQuery(\n",
    "                    chemical_formula=ChemicalFieldQuery(\n",
    "                        extract_as=\"formula\",\n",
    "                        filter=ChemicalFilter(equal='AlxCuy'),\n",
    "                    ),\n",
    "                    properties=PropertyQuery(\n",
    "                        name=FieldQuery(\n",
    "                            filter=[Filter(equal=\"Formation energy\"), Filter(equal=\"Enthalpy of Formation\")]),\n",
    "                        value=FieldQuery(\n",
    "                            extract_as=\"formation_enthalpy\")\n",
    "                    )\n",
    "                )\n",
    "            ))\n",
    "\n",
    "res = client.search.pif_search(query)\n",
    "print(\"Found {} PIFs\".format(res.total_num_hits))\n",
    "print([x.extracted for x in res.hits][0:2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "A few notes:\n",
    " * `extract_as` extracts the field it is on (why we need value)\n",
    " * `extracted` is a dictionary from String keys to values; no units or uncertainty"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Real example: phase stability diagram\n",
    "\n",
    "Let's create an Al-Cu phase diagram and add these points."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "We'll start with some sample DFT data and create pifs out of the samples. Then, we'll add the new points found in our query above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from dfttopif import directory_to_pif\n",
    "\n",
    "Cu_pif = directory_to_pif(\"./example_data/Cu.cF4\")\n",
    "Al_pif = directory_to_pif(\"./example_data/Al.cF4\")\n",
    "AlCu_pifs = [directory_to_pif(os.path.join(\"./example_data/\", x))\n",
    "             for x in os.listdir(\"./example_data/\") if \"Al\" in x]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def enthalpy_of_formation(energy, n_Al, n_Cu, energy_Al, energy_Cu):\n",
    "    return (energy - n_Al * energy_Al - n_Cu * energy_Cu) / (n_Al + n_Cu)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "\n",
    "def get_stoich(AlCu_formula):\n",
    "    m = re.match(r\"Al([0-9]*)Cu([0-9]*)\", AlCu_formula)\n",
    "    if m is not None:\n",
    "        n_Al = float(m.group(1)) if len(m.group(1)) > 0 else 1\n",
    "        n_Cu = float(m.group(2)) if len(m.group(2)) > 0 else 1\n",
    "        return (n_Al, n_Cu)\n",
    "    m = re.match(r\"Cu([0-9]*)Al([0-9]*)\", AlCu_formula)\n",
    "    if m is not None:\n",
    "        n_Cu = float(m.group(1)) if len(m.group(1)) > 0 else 1\n",
    "        n_Al = float(m.group(2)) if len(m.group(2)) > 0 else 1\n",
    "        return (n_Al, n_Cu)\n",
    "    return (0, 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ReadView is a dictionary-like read-only view into a PIF that indexes members by name and elevates keys up the PIF hierarchy as long as they are unambiguous. For more info see the docs.\n",
    "\n",
    "Here, we'll use ReadView to populate an array of `(Energy, formula)` tuples to build our phase diagram."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pypif_sdk.readview import ReadView\n",
    "\n",
    "energy_Al = ReadView(Al_pif)[\"Total Energy\"].scalars[0].value / 4\n",
    "energy_Cu = ReadView(Cu_pif)[\"Total Energy\"].scalars[0].value\n",
    "\n",
    "points = [(0.0, 0.0), (1.0, 0.0)]\n",
    "for pif in AlCu_pifs:\n",
    "    rv = ReadView(pif)\n",
    "    energy = rv[\"Total Energy\"].scalars[0].value\n",
    "    n_Al, n_Cu = get_stoich(rv.chemical_formula)\n",
    "    if n_Al == 0 and n_Cu == 0: continue\n",
    "    points.append((\n",
    "            n_Cu / (n_Cu + n_Al),\n",
    "            enthalpy_of_formation(energy, n_Al, n_Cu, energy_Al, energy_Cu)\n",
    "        ))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "Now, let's add the new points from our queries above. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "new_points = []\n",
    "for hit in res.hits:\n",
    "    n_Al, n_Cu = get_stoich(hit.extracted[\"formula\"])\n",
    "    new_points.append((\n",
    "            n_Cu / (n_Cu + n_Al),\n",
    "            float(hit.extracted[\"formation_enthalpy\"])\n",
    "        ))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'$\\\\Delta H (eV/atom)$')"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x111a5c5f8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.scatter(*zip(*points), color='blue', marker='o', label='alloyDB')\n",
    "plt.scatter(*zip(*new_points), color='red', marker='x', label='citrination')\n",
    "plt.legend(loc=\"best\")\n",
    "plt.xlim(0, 1)\n",
    "plt.xlabel(\"Cu fraction\")\n",
    "plt.ylabel(\"$\\Delta H (eV/atom)$\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "### Filter by value\n",
    "\n",
    "A few of the values in are mislabeled: they are called \"Enthalpy of Formation\" but are actually total energies (per atom).  We can filter them out by recognizing that all the Al-Cu binaries have enthalpy of formation above $-1.0 eV / $atom."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "query = PifSystemReturningQuery(\n",
    "            size=500,\n",
    "            query=DataQuery(\n",
    "                system=PifSystemQuery(\n",
    "                    chemical_formula=ChemicalFieldQuery(\n",
    "                        extract_as=\"formula\",\n",
    "                        filter=ChemicalFilter(equal='AlxCuy'),\n",
    "                    ),\n",
    "                    properties=PropertyQuery(\n",
    "                        name=FieldQuery(\n",
    "                            filter=[Filter(equal=\"Formation energy\"), Filter(equal=\"Enthalpy of Formation\")]),\n",
    "                        value=FieldQuery(\n",
    "                            extract_as=\"formation_enthalpy\",\n",
    "                            filter=[Filter(min=-1.0)])\n",
    "                    )\n",
    "                )\n",
    "            ))\n",
    "\n",
    "query_result = client.search.pif_search(query)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "new_points = []\n",
    "for hit in query_result.hits:\n",
    "    n_Al, n_Cu = get_stoich(hit.extracted[\"formula\"])\n",
    "    new_points.append((\n",
    "            n_Cu / (n_Cu + n_Al),\n",
    "            float(hit.extracted[\"formation_enthalpy\"])\n",
    "        ))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "This looks much better:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'$\\\\Delta H (eV/atom)$')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10b129b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "plt.scatter(*zip(*points), color='blue', marker='o', label='alloyDB')\n",
    "plt.scatter(*zip(*new_points), color='red', marker='x', label='citrination (filtered)')\n",
    "plt.legend()\n",
    "plt.xlim(0, 1)\n",
    "plt.xlabel(\"Cu fraction\")\n",
    "plt.ylabel(\"$\\Delta H (eV/atom)$\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "celltoolbar": "Slideshow",
  "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": "3.6.4"
  },
  "nbpresent": {
   "slides": {
    "5c8082ad-f7f3-497a-91a5-b82f1bfa6d0e": {
     "id": "5c8082ad-f7f3-497a-91a5-b82f1bfa6d0e",
     "layout": "grid",
     "prev": null,
     "regions": {
      "3fdabdaa-d1a7-4289-8161-39096a883c62": {
       "attrs": {
        "height": 1,
        "pad": 0.01,
        "treemap:weight": 1,
        "width": 1,
        "x": 0,
        "y": 0
       },
       "content": {
        "cell": "f2f733dc-95c8-4751-a337-7bba7fcbe4d7",
        "part": "whole"
       },
       "id": "3fdabdaa-d1a7-4289-8161-39096a883c62"
      }
     }
    },
    "8e99069f-4d09-4e7b-8dbb-b8cf82a67ae2": {
     "id": "8e99069f-4d09-4e7b-8dbb-b8cf82a67ae2",
     "prev": "5c8082ad-f7f3-497a-91a5-b82f1bfa6d0e",
     "regions": {
      "1dabf2fc-3b38-4a0d-a95c-c4574fd5354f": {
       "attrs": {
        "height": 0.6,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "c41a4f53-9a6a-435e-b554-90780aa20fcd",
        "part": "whole"
       },
       "id": "1dabf2fc-3b38-4a0d-a95c-c4574fd5354f"
      },
      "ab4f31d7-8029-4c39-854f-edaad678b711": {
       "attrs": {
        "height": 0.2,
        "width": 0.4,
        "x": 0.5,
        "y": 0.7
       },
       "id": "ab4f31d7-8029-4c39-854f-edaad678b711"
      },
      "b3bd2bc4-2c69-4998-8fe0-a09a6c579492": {
       "attrs": {
        "height": 0.2,
        "width": 0.4,
        "x": 0.1,
        "y": 0.7
       },
       "id": "b3bd2bc4-2c69-4998-8fe0-a09a6c579492"
      }
     }
    }
   },
   "themes": {}
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}