tsneDemo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Citrination t-SNE API"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this tutorial, we're going to learn to use Citrination to generate a useful data analysis called t-SNE. This data visualization technique enables you to represent a high dimensional set of data in fewer dimensions in a way that preserves the local structure of the data. In materials informatics, this allows you to create a two-dimensional plot of a set of materials where points corresponding to similar materials are grouped together in two-dimensional space. More information on t-SNE here: https://lvdmaaten.github.io/tsne/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This tutorial will teach you to create and export a two-dimensional t-SNE plot for any data on Citrination. The first step is to create a data view on the Citrination: https://citrination.com/data_views. Instructions for creating a data view can be found in [this tutorial](https://github.com/CitrineInformatics/learn-citrination/blob/master/Journal%20Paper%20to%20Model%20Demo.ipynb).\n",
    "\n",
    "We'll be using this data view: https://citrination.com/data_views/787 (view id 787) for this tutorial, which includes a model predicting experimental band gaps based on data compiled by W.H. Strehlow and E.L. Cook, which can be viewed in [this dataset](https://citrination.com/datasets/1160/show_search)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from citrination_client import CitrinationClient\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['Property Band gap', 'Property Color'])\n"
     ]
    }
   ],
   "source": [
    "client = CitrinationClient(api_key=os.environ[\"CITRINATION_API_KEY\"])\n",
    "\n",
    "tsne_dataviewid = \"4106\"\n",
    "tsne = client.models.tsne(tsne_dataviewid)\n",
    "print(tsne.projections())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `tsne` call returns a dictionary where the top level keys correspond to the properties that are being modeled. In this data view, the model inputs are Chemical formula and Crystallinity. Citrine represents a chemical formula by calculating over 50 material descriptors. The t-SNE plot reduces that into just 2 dimensions. Let's look at the t-SNE plot for Band gap."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<citrination_client.models.projection.Projection object at 0x10f197c50>\n"
     ]
    }
   ],
   "source": [
    "print(tsne.get_projection(\"Property Band gap\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- `x` is the x component of the t-SNE plot\n",
    "- `y` is the y component of the t-SNE plot\n",
    "- `z` is the property value (Band gap in this example)\n",
    "- `uid` is the uid of the record on Citrination. You can find the record by going to https://citrination.com/pif/{uid}\n",
    "- `label` is the list of the values of the model inputs for the record"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/seto/anaconda2/envs/tensorflow36/lib/python3.6/site-packages/matplotlib/cbook/deprecation.py:106: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  warnings.warn(message, mplDeprecation, stacklevel=1)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1103f8c18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gap_projection = tsne.get_projection(\"Property Band gap\")\n",
    "plt.scatter(gap_projection.xs, gap_projection.ys, c=gap_projection.responses)\n",
    "plt.colorbar(label=\"Band gap (eV)\")\n",
    "plt.xlabel(\"t-SNE Component 1\")\n",
    "plt.ylabel(\"t-SNE Component 2\")\n",
    "plt.axes().set_aspect('equal', 'datalim')\n",
    "# plt.xlim(-2500,2500)\n",
    "# plt.ylim(-500,3500)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When interpreting a t-SNE plot, keep in mind that the t-SNE components plotted on the x and y axes have no physical meaning. Since the t-SNE algorithm preserves the local structure of the data, we look at points that end up close together. Just from looking at the plot, we can see that there is a cluster of high bandgap materials that contains the material with the maximum band gap in the bottom right corner. Let's take a look at what those materials are.\n",
    "\n",
    "\n",
    "(note: t-SNE is a non-deterministic algorithm, so the plot will change if the underlying Data View is retrained)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the coordinates and band gap values into a numpy array\n",
    "coordinates_and_bg = np.zeros((len(gap_projection.xs), 3))\n",
    "coordinates_and_bg[:,0] = gap_projection.xs\n",
    "coordinates_and_bg[:,1] = gap_projection.ys\n",
    "coordinates_and_bg[:,2] = gap_projection.responses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find the n materials closest to the target material, as specified by index\n",
    "# based on cartesian distance in the t-SNE coordinates\n",
    "def find_most_similar_materials(index, coordinates, n_materials=10):\n",
    "    distances = np.linalg.norm(coordinates_and_bg[:,0:2] - coordinates_and_bg[index,0:2], axis=1)\n",
    "    return np.argsort(distances)[1:n_materials+1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From inspecting this cluster of materials, we can see that the t-SNE plot has grouped highly ionic materials together, and that these materials have very high band gaps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Maximum band gap material:\n",
      "Be1O1, Band gap: 14.5, URL: https://citrination.com/pif/1160/3/66756141934A40C0096E45068CE9B071\n",
      "Similar materials:\n",
      "B2O3, Polycrystalline, Band gap: 4.9, URL: https://citrination.com/pif/1160/3/C98CE9CE4C27BA625CF3478677A2D0C5\n",
      "V1O2, Single crystalline, Band gap: 0.65, URL: https://citrination.com/pif/1160/3/DA5FA80CB61BDD6A092075F6817E42FD\n",
      "V2O3, Single crystalline, Band gap: 0.1, URL: https://citrination.com/pif/1160/3/E694087444A1F8F0BD04E8934EE1A5F7\n",
      "Ti1O2, Single crystalline, Band gap: 3.05, URL: https://citrination.com/pif/1160/3/F734845B82D2DEE1CB882ECEB37AEEC6\n",
      "V1O1, Band gap: 0.3, URL: https://citrination.com/pif/1160/3/13A857F2BEA1AF21C9495CA6BFEE37BF\n",
      "V4O7, Single crystalline, Band gap: 0.8, URL: https://citrination.com/pif/1160/3/157376D5171A19965C6D6218C8B4F01E\n",
      "V2O5, Single crystalline, Band gap: 2.54, URL: https://citrination.com/pif/1160/3/CA4ED13625FFF3FFD77AA0446116B87B\n",
      "Mg1O1, Single crystalline, Band gap: 7.69, URL: https://citrination.com/pif/1160/3/0A5A69057521F06C37687B8AAE2E0905\n",
      "Cr1O2, Polycrystalline, Band gap: 0.23, URL: https://citrination.com/pif/1160/3/D45EB1BFD5F6EBEBD7E7D8BF3A43862C\n",
      "Zr1O2, Band gap: 0.65, URL: https://citrination.com/pif/1160/3/C505DB1DA4187DA22A3252185B5335D9\n",
      "-651.6577729506878 49984.637839736446 -13976.535209245418 573.5589153283767\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11053a9b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the index of the maximum band gap material\n",
    "max_bg_index = np.argmax(coordinates_and_bg, axis=0)[2]\n",
    "print(\"Maximum band gap material:\")\n",
    "print(\"{}, Band gap: {}, URL: https://citrination.com/pif/{}\"\n",
    "      .format(gap_projection.tags[max_bg_index], \n",
    "              gap_projection.responses[max_bg_index],\n",
    "              gap_projection.uids[max_bg_index])\n",
    "     )\n",
    "print(\"Similar materials:\")\n",
    "for idx in find_most_similar_materials(max_bg_index, coordinates_and_bg[:,0:2]):\n",
    "    print(\"{}, Band gap: {}, URL: https://citrination.com/pif/{}\"\n",
    "          .format(gap_projection.tags[idx],\n",
    "                  gap_projection.responses[idx],\n",
    "                  gap_projection.uids[idx])\n",
    "         )\n",
    "\n",
    "plt.figure()\n",
    "plt.scatter(gap_projection.xs, gap_projection.ys, c=gap_projection.responses)\n",
    "plt.colorbar(label=\"Band gap (eV)\")\n",
    "plt.scatter(coordinates_and_bg[max_bg_index, 0], coordinates_and_bg[max_bg_index, 1],\n",
    "            marker=\"*\", c=\"r\", s=200)\n",
    "plt.xlabel(\"t-SNE Component 1\")\n",
    "plt.ylabel(\"t-SNE Component 2\")\n",
    "#plt.axes().set_aspect('equal', 'datalim')\n",
    "print(min(gap_projection.xs), max(gap_projection.xs), min(gap_projection.ys), max(gap_projection.ys))\n",
    "plt.xlim(-5000,55000)\n",
    "plt.ylim(-20000,6000)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's figure out what material the outlier point with the largest y-value corresponds to"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Outlier material (point with the highest y-value):\n",
      "La2S3, Polycrystalline, Band gap: 1.32, URL: https://citrination.com/pif/1160/3/029F203BCD1C63011185C262945FD909\n",
      "-651.6577729506878 49984.637839736446 -13976.535209245418 573.5589153283767\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/seto/anaconda2/envs/tensorflow36/lib/python3.6/site-packages/matplotlib/cbook/deprecation.py:106: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  warnings.warn(message, mplDeprecation, stacklevel=1)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x111141160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "outlier_index = np.argmax(coordinates_and_bg, axis=0)[1]\n",
    "print(\"Outlier material (point with the highest y-value):\")\n",
    "print(\"{}, Band gap: {}, URL: https://citrination.com/pif/{}\"\n",
    "      .format(gap_projection.tags[outlier_index],\n",
    "              gap_projection.responses[outlier_index],\n",
    "              gap_projection.uids[outlier_index])\n",
    "     )\n",
    "\n",
    "plt.figure()\n",
    "plt.scatter(gap_projection.xs, gap_projection.ys, c=gap_projection.responses)\n",
    "plt.colorbar(label=\"Band gap (eV)\")\n",
    "plt.scatter(coordinates_and_bg[outlier_index, 0], coordinates_and_bg[outlier_index, 1],\n",
    "            marker=\"*\", c=\"r\", s=200)\n",
    "plt.xlabel(\"t-SNE Component 1\")\n",
    "plt.ylabel(\"t-SNE Component 2\")\n",
    "print(min(gap_projection.xs), max(gap_projection.xs), min(gap_projection.ys), max(gap_projection.ys))\n",
    "plt.xlim(-5000,55000)\n",
    "plt.ylim(-20000,6000)\n",
    "plt.axes().set_aspect('equal', 'datalim')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also find materials that are similar to a given material in the dataset. Let's find materials similar to Cr3Se4 https://citrination.com/datasets/1160/version/3/pif/0478C7E0B7C4F75C6081EC5184522704. The uid comes in the format dataset_id/version/pif_id. For Cr3Se4, the uid is 1160/3/0478C7E0B7C4F75C6081EC5184522704."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cr3Se4, Band gap: 0.015, URL: https://citrination.com/pif/1160/3/0478C7E0B7C4F75C6081EC5184522704\n",
      "Similar materials:\n",
      "Cr2Se3, Band gap: 0.15, URL: https://citrination.com/pif/1160/3/B02503AE0620DB98FC19B2637224AAEC\n",
      "Mn1Se1, Single crystalline, Band gap: 1.8, URL: https://citrination.com/pif/1160/3/57C264FAE60E73465FA484FCC686CFD8\n",
      "Ru1Se2, Polycrystalline, Band gap: 1.0, URL: https://citrination.com/pif/1160/3/EB7F89526BB12CD5DB57956267A4B814\n",
      "Fe1Se2, Polycrystalline, Band gap: 0.5, URL: https://citrination.com/pif/1160/3/5FE698B58A042D23AA8B6EF0CB7C4923\n",
      "Rh1Se3, Band gap: 0.7, URL: https://citrination.com/pif/1160/3/30EEC951791B144F58B004220982F02B\n",
      "Rh1Se2, Band gap: 0.6, URL: https://citrination.com/pif/1160/3/53AE437CD930A984FD4C45C29F2F1B5F\n",
      "Ru1P2, Polycrystalline, Band gap: 1.0, URL: https://citrination.com/pif/1160/3/532047FC884C8662B6F01D9759035CF6\n",
      "Rh2S3, Band gap: 0.8, URL: https://citrination.com/pif/1160/3/5663B427D1ADBB8DBA1B8B503E394785\n",
      "W1S2, Single crystalline, Band gap: 0.45, URL: https://citrination.com/pif/1160/3/C685CC068EAA30FF9DD3C8F27E77C42C\n",
      "Mn1Se2, Polycrystalline, Band gap: 0.2, URL: https://citrination.com/pif/1160/3/FC26DD83901588E628E6F0FFE5878CCA\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/seto/anaconda2/envs/tensorflow36/lib/python3.6/site-packages/matplotlib/cbook/deprecation.py:106: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  warnings.warn(message, mplDeprecation, stacklevel=1)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10f104978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "uid = \"1160/3/0478C7E0B7C4F75C6081EC5184522704\"\n",
    "cr3se4_index = gap_projection.uids.index(uid)\n",
    "print(\"{}, Band gap: {}, URL: https://citrination.com/pif/{}\"\n",
    "      .format(gap_projection.tags[cr3se4_index],\n",
    "              gap_projection.responses[cr3se4_index],\n",
    "              gap_projection.uids[cr3se4_index])\n",
    "     )\n",
    "print(\"Similar materials:\")\n",
    "for idx in find_most_similar_materials(cr3se4_index, coordinates_and_bg[:,0:2]):\n",
    "    print(\"{}, Band gap: {}, URL: https://citrination.com/pif/{}\"\n",
    "          .format(gap_projection.tags[idx],\n",
    "                  gap_projection.responses[idx],\n",
    "                  gap_projection.uids[idx])\n",
    "         )\n",
    "plt.figure()\n",
    "plt.scatter(gap_projection.xs, gap_projection.ys, c=gap_projection.responses)\n",
    "plt.colorbar(label=\"Band gap (eV)\")\n",
    "plt.scatter(coordinates_and_bg[cr3se4_index, 0], coordinates_and_bg[cr3se4_index, 1],\n",
    "            marker=\"*\", c=\"r\", s=200)\n",
    "plt.xlabel(\"t-SNE Component 1\")\n",
    "plt.ylabel(\"t-SNE Component 2\")\n",
    "plt.axes().set_aspect('equal', 'datalim')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we've seen here, t-SNE is a useful tool for understanding materials datasets and gaining new insights by visualizing and exploring a large, high-dimensional dataset in two dimensions. In this tutorial, we've seen how to examine outliers, find interesting clusters of similar materials, and find materials that are similar to a given material, which are only some of the cool things that you can do with t-SNE plots. The t-SNE analysis is available for all real-valued outputs specified in the Data Views creation process."
   ]
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}