Skills-ML Tour.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Skills-ML Tour\n",
    "\n",
    "Skills-ML is an open source software library for applying NLP and ML to labor market data. It allows the user to perform tasks like skill extraction and occupation classification to collections of documents such as job postings, profiles, and course descriptions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Competency\n",
    "\n",
    "A competency is any expertise or talent that is useful for a job. Developed capacities (e.g. active listening), proficiency with tools or technology (e.g. lancets, Microsoft Word), innate abilities (e.g. originality), and academic knowledge (e.g. medicine) are all considered competencies."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.ontologies import Competency\n",
    "\n",
    "dinosaur_riding = Competency(\n",
    "    identifier='dino_riding',\n",
    "    name='Dinosaur Riding',\n",
    "    description='Using the back of a dinosaur for transportation'\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Competency Relationships\n",
    "\n",
    "Competencies are often related to each other. Defining parent-child relationships is a standard building block of existing competency frameworks like ONET and ESCO. A parent-child relationship generally implies that the child is a \"type of\" the parent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.38.0 (20140413.2041)\n",
       " -->\n",
       "<!-- Title: %3 Pages: 1 -->\n",
       "<svg width=\"519pt\" height=\"116pt\"\n",
       " viewBox=\"0.00 0.00 519.33 116.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 112)\">\n",
       "<title>%3</title>\n",
       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-112 515.334,-112 515.334,4 -4,4\"/>\n",
       "<!-- 12345 -->\n",
       "<g id=\"node1\" class=\"node\"><title>12345</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"85.7924\" cy=\"-18\" rx=\"85.5853\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"85.7924\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">Dinosaur Riding</text>\n",
       "</g>\n",
       "<!-- 12346 -->\n",
       "<g id=\"node2\" class=\"node\"><title>12346</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"260.792\" cy=\"-18\" rx=\"71.4873\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"260.792\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">Train Surfing</text>\n",
       "</g>\n",
       "<!-- 123 -->\n",
       "<g id=\"node3\" class=\"node\"><title>123</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"132.792\" cy=\"-90\" rx=\"122.678\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"132.792\" y=\"-86.3\" font-family=\"Times,serif\" font-size=\"14.00\">Extreme Transportation</text>\n",
       "</g>\n",
       "<!-- 123&#45;&#45;12345 -->\n",
       "<g id=\"edge1\" class=\"edge\"><title>123&#45;&#45;12345</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M121.415,-72.055C114.134,-61.2111 104.715,-47.183 97.3915,-36.2753\"/>\n",
       "</g>\n",
       "<!-- 123&#45;&#45;12346 -->\n",
       "<g id=\"edge2\" class=\"edge\"><title>123&#45;&#45;12346</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M163.127,-72.411C183.974,-61.0098 211.485,-45.965 231.994,-34.7489\"/>\n",
       "</g>\n",
       "<!-- 23456 -->\n",
       "<g id=\"node4\" class=\"node\"><title>23456</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"415.792\" cy=\"-18\" rx=\"65.7887\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"415.792\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">Time Travel</text>\n",
       "</g>\n",
       "<!-- 2345 -->\n",
       "<g id=\"node5\" class=\"node\"><title>2345</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"415.792\" cy=\"-90\" rx=\"95.5831\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"415.792\" y=\"-86.3\" font-family=\"Times,serif\" font-size=\"14.00\">Advanced Science</text>\n",
       "</g>\n",
       "<!-- 2345&#45;&#45;23456 -->\n",
       "<g id=\"edge3\" class=\"edge\"><title>2345&#45;&#45;23456</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M415.792,-71.6966C415.792,-60.8463 415.792,-46.9167 415.792,-36.1043\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.dot.Graph at 0x7f573433fac8>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from skills_ml.ontologies import Competency\n",
    "from skills_ml.ontologies.viz import display_nodes\n",
    "\n",
    "dinosaur_riding = Competency(\n",
    "    identifier='12345',\n",
    "    name='Dinosaur Riding',\n",
    "    description='Using the back of a dinosaur for transportation'\n",
    ")\n",
    "\n",
    "train_surfing = Competency(\n",
    "    identifier='12346',\n",
    "    name='Train Surfing',\n",
    "    description='Standing on the train while it goes'\n",
    ")\n",
    "\n",
    "time_travel = Competency(\n",
    "    identifier='23456',\n",
    "    name='Time Travel',\n",
    "    description='Traveling Through Time'\n",
    ")\n",
    "\n",
    "advanced_science = Competency(\n",
    "    identifier='2345',\n",
    "    name='Advanced Science',\n",
    ")\n",
    "\n",
    "extreme_transportation = Competency(\n",
    "    identifier='123',\n",
    "    name='Extreme Transportation',\n",
    "    description='Comically dangerous forms of transportation'\n",
    ")\n",
    "time_travel.add_parent(advanced_science)\n",
    "dinosaur_riding.add_parent(extreme_transportation)\n",
    "train_surfing.add_parent(extreme_transportation)\n",
    "display_nodes([dinosaur_riding, train_surfing, extreme_transportation, time_travel, advanced_science])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Occupation\n",
    "\n",
    "An occupation is a job or profession that a person can hold. Similar to competencies, these are also often defined hierarchically.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.38.0 (20140413.2041)\n",
       " -->\n",
       "<!-- Title: %3 Pages: 1 -->\n",
       "<svg width=\"652pt\" height=\"116pt\"\n",
       " viewBox=\"0.00 0.00 652.27 116.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 112)\">\n",
       "<title>%3</title>\n",
       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-112 648.272,-112 648.272,4 -4,4\"/>\n",
       "<!-- 999 -->\n",
       "<g id=\"node1\" class=\"node\"><title>999</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"313.837\" cy=\"-90\" rx=\"124.278\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"313.837\" y=\"-86.3\" font-family=\"Times,serif\" font-size=\"14.00\">Extreme Postal Workers</text>\n",
       "</g>\n",
       "<!-- 9998 -->\n",
       "<g id=\"node2\" class=\"node\"><title>9998</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"148.837\" cy=\"-18\" rx=\"148.674\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"148.837\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">Deliverer of Items to the Past</text>\n",
       "</g>\n",
       "<!-- 999&#45;&#45;9998 -->\n",
       "<g id=\"edge1\" class=\"edge\"><title>999&#45;&#45;9998</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M275.568,-72.7646C249.102,-61.5368 214.014,-46.651 187.47,-35.3898\"/>\n",
       "</g>\n",
       "<!-- 9999 -->\n",
       "<g id=\"node3\" class=\"node\"><title>9999</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"479.837\" cy=\"-18\" rx=\"164.371\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"479.837\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">Deliverer of Items to Train Yards</text>\n",
       "</g>\n",
       "<!-- 999&#45;&#45;9999 -->\n",
       "<g id=\"edge2\" class=\"edge\"><title>999&#45;&#45;9999</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M352.338,-72.7646C378.868,-61.5771 414.011,-46.7581 440.681,-35.5115\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.dot.Graph at 0x7f573433f588>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from skills_ml.ontologies import Occupation\n",
    "\n",
    "extreme_postal_workers = Occupation(identifier='999', name='Extreme Postal Workers')\n",
    "dino_postal_worker = Occupation(identifier='9998', name='Deliverer of Items to the Past')\n",
    "train_yard_postal_worker = Occupation(identifier='9999', name='Deliverer of Items to Train Yards')\n",
    "dino_postal_worker.add_parent(extreme_postal_workers)\n",
    "train_yard_postal_worker.add_parent(extreme_postal_workers)\n",
    "\n",
    "display_nodes([extreme_postal_workers, dino_postal_worker, train_yard_postal_worker])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CompetencyOntology\n",
    "\n",
    "A CompetencyOntology is a model of the labor market, or some subset thereof, consisting of a collection of competencies, a collection of occupations, and all of the relationships between them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.38.0 (20140413.2041)\n",
       " -->\n",
       "<!-- Title: %3 Pages: 1 -->\n",
       "<svg width=\"1189pt\" height=\"246pt\"\n",
       " viewBox=\"0.00 0.00 1189.00 246.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 242)\">\n",
       "<title>%3</title>\n",
       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-242 1185,-242 1185,4 -4,4\"/>\n",
       "<g id=\"clust1\" class=\"cluster\"><title>cluster_competencies</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"8,-80 8,-230 505,-230 505,-80 8,-80\"/>\n",
       "<text text-anchor=\"middle\" x=\"256.5\" y=\"-214.8\" font-family=\"Times,serif\" font-size=\"14.00\">Competencies</text>\n",
       "</g>\n",
       "<g id=\"clust2\" class=\"cluster\"><title>cluster_occupations</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"513,-8 513,-155 1173,-155 1173,-8 513,-8\"/>\n",
       "<text text-anchor=\"middle\" x=\"843\" y=\"-139.8\" font-family=\"Times,serif\" font-size=\"14.00\">Occupations</text>\n",
       "</g>\n",
       "<!-- 123 -->\n",
       "<g id=\"node1\" class=\"node\"><title>123</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"361\" cy=\"-181\" rx=\"122.678\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"361\" y=\"-177.3\" font-family=\"Times,serif\" font-size=\"14.00\">Extreme Transportation</text>\n",
       "</g>\n",
       "<!-- 12345 -->\n",
       "<g id=\"node2\" class=\"node\"><title>12345</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"251\" cy=\"-106\" rx=\"85.5853\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"251\" y=\"-102.3\" font-family=\"Times,serif\" font-size=\"14.00\">Dinosaur Riding</text>\n",
       "</g>\n",
       "<!-- 123&#45;&#45;12345 -->\n",
       "<g id=\"edge1\" class=\"edge\"><title>123&#45;&#45;12345</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M335.764,-163.252C317.75,-151.298 293.64,-135.298 275.74,-123.418\"/>\n",
       "</g>\n",
       "<!-- 12346 -->\n",
       "<g id=\"node3\" class=\"node\"><title>12346</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"426\" cy=\"-106\" rx=\"71.4873\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"426\" y=\"-102.3\" font-family=\"Times,serif\" font-size=\"14.00\">Train Surfing</text>\n",
       "</g>\n",
       "<!-- 123&#45;&#45;12346 -->\n",
       "<g id=\"edge2\" class=\"edge\"><title>123&#45;&#45;12346</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M376.076,-163.069C386.577,-151.276 400.523,-135.613 411.006,-123.839\"/>\n",
       "</g>\n",
       "<!-- 9998 -->\n",
       "<g id=\"node6\" class=\"node\"><title>9998</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"670\" cy=\"-34\" rx=\"148.674\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"670\" y=\"-30.3\" font-family=\"Times,serif\" font-size=\"14.00\">Deliverer of Items to the Past</text>\n",
       "</g>\n",
       "<!-- 12345&#45;&#45;9998 -->\n",
       "<g id=\"edge8\" class=\"edge\"><title>12345&#45;&#45;9998</title>\n",
       "<path fill=\"none\" stroke=\"black\" stroke-dasharray=\"5,2\" d=\"M299.557,-91.0293C314.38,-87.0877 330.783,-83.0478 346,-80 417.972,-65.5855 499.859,-54.2555 563.303,-46.5697\"/>\n",
       "</g>\n",
       "<!-- 9999 -->\n",
       "<g id=\"node7\" class=\"node\"><title>9999</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"1001\" cy=\"-34\" rx=\"164.371\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"1001\" y=\"-30.3\" font-family=\"Times,serif\" font-size=\"14.00\">Deliverer of Items to Train Yards</text>\n",
       "</g>\n",
       "<!-- 12346&#45;&#45;9999 -->\n",
       "<g id=\"edge7\" class=\"edge\"><title>12346&#45;&#45;9999</title>\n",
       "<path fill=\"none\" stroke=\"black\" stroke-dasharray=\"5,2\" d=\"M466.535,-90.9779C479.911,-86.8399 494.961,-82.7008 509,-80 648.76,-53.1133 686.289,-65.1816 828,-52 846.03,-50.3229 865.111,-48.4981 883.761,-46.687\"/>\n",
       "</g>\n",
       "<!-- 23456 -->\n",
       "<g id=\"node4\" class=\"node\"><title>23456</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"82\" cy=\"-106\" rx=\"65.7887\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"82\" y=\"-102.3\" font-family=\"Times,serif\" font-size=\"14.00\">Time Travel</text>\n",
       "</g>\n",
       "<!-- 23456&#45;&#45;9998 -->\n",
       "<g id=\"edge6\" class=\"edge\"><title>23456&#45;&#45;9998</title>\n",
       "<path fill=\"none\" stroke=\"black\" stroke-dasharray=\"5,2\" d=\"M118.031,-90.9082C129.95,-86.7661 143.391,-82.6422 156,-80 279.181,-54.1873 421.975,-43.1535 525.749,-38.4516\"/>\n",
       "</g>\n",
       "<!-- 2345 -->\n",
       "<g id=\"node5\" class=\"node\"><title>2345</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"112\" cy=\"-181\" rx=\"95.5831\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"112\" y=\"-177.3\" font-family=\"Times,serif\" font-size=\"14.00\">Advanced Science</text>\n",
       "</g>\n",
       "<!-- 2345&#45;&#45;23456 -->\n",
       "<g id=\"edge3\" class=\"edge\"><title>2345&#45;&#45;23456</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M104.89,-162.7C100.074,-150.979 93.739,-135.565 88.9551,-123.924\"/>\n",
       "</g>\n",
       "<!-- 999 -->\n",
       "<g id=\"node8\" class=\"node\"><title>999</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"835\" cy=\"-106\" rx=\"124.278\" ry=\"18\"/>\n",
       "<text text-anchor=\"middle\" x=\"835\" y=\"-102.3\" font-family=\"Times,serif\" font-size=\"14.00\">Extreme Postal Workers</text>\n",
       "</g>\n",
       "<!-- 999&#45;&#45;9998 -->\n",
       "<g id=\"edge4\" class=\"edge\"><title>999&#45;&#45;9998</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M796.731,-88.7646C770.265,-77.5368 735.177,-62.651 708.633,-51.3898\"/>\n",
       "</g>\n",
       "<!-- 999&#45;&#45;9999 -->\n",
       "<g id=\"edge5\" class=\"edge\"><title>999&#45;&#45;9999</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M873.501,-88.7646C900.031,-77.5771 935.174,-62.7581 961.844,-51.5115\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.dot.Graph at 0x7f57342fe160>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from skills_ml.ontologies import CompetencyOntology\n",
    "from skills_ml.ontologies.viz import display_ontology\n",
    "\n",
    "ontology = CompetencyOntology()\n",
    "ontology.add_competency(dinosaur_riding)\n",
    "ontology.add_competency(train_surfing)\n",
    "ontology.add_competency(extreme_transportation)\n",
    "ontology.add_competency(time_travel)\n",
    "ontology.add_competency(advanced_science)\n",
    "\n",
    "ontology.add_occupation(dino_postal_worker)\n",
    "ontology.add_occupation(train_yard_postal_worker)\n",
    "ontology.add_occupation(extreme_postal_workers)\n",
    "\n",
    "ontology.add_edge(occupation=dino_postal_worker, competency=dinosaur_riding)\n",
    "ontology.add_edge(occupation=dino_postal_worker, competency=time_travel)\n",
    "ontology.add_edge(occupation=train_yard_postal_worker, competency=train_surfing)\n",
    "\n",
    "display_ontology(ontology)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prebuilt Ontologies\n",
    "\n",
    "To move on we'll want to level up to a full ontology. The example we'll use is O\\*NET, built from survey data and maintained by the US Department of Labor. A CompetencyOntology subclass that downloads the source files from the O\\*NET web site is included in Skills-ML."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ontology summary statistics for onet\n",
      "Num competencies: 32030\n",
      "Num occupations: 1133\n",
      "Num competency-occupation edges: 107305\n",
      "Median occupations per competency: 1\n",
      "Median competencies per occupation: 89\n",
      "Mean occupations per competency: 3.350245090386837\n",
      "Mean competencies per occupation: 94.70873786407768\n"
     ]
    }
   ],
   "source": [
    "from skills_ml.ontologies.onet import Onet\n",
    "onet = Onet()\n",
    "onet.print_summary_stats()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Competency(identifier=43232108-Rezgo, name=Rezgo, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232605-Hansen CAMA, name=Hansen CAMA, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=27111526-Arrow squaring devices, name=Arrow squaring devices, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=48101817-Marzipan tools, name=Marzipan tools, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=21111507-Commercial fishing line sinkers, name=Commercial fishing line sinkers, categories=['Tools', 'O*NET T2'], {})]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(onet.competencies)[0:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Filtering\n",
    "\n",
    "~34000 competencies and ~1100 occupations is a lot. Let's explore the filtering functionality of the CompetencyOntology to zoom in on a more specific slice.\n",
    "\n",
    "`filter_by` filters using edges: the filtering function it expects takes in an edge (between a Competency and Occupation) and returns whether or not it should be in the result. The result takes the form of a new CompetencyOntology, so you can interact with it in the same way as you would the source ontology. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{Competency(identifier=1.A.1.a.1, name=Oral Comprehension, categories=['Abilities'], {'competencyText': 'The ability to listen to and understand information and ideas presented through spoken words and sentences.'}),\n",
       " Competency(identifier=1.A.1.a.2, name=Written Comprehension, categories=['Abilities'], {'competencyText': 'The ability to read and understand information and ideas presented in writing.'}),\n",
       " Competency(identifier=1.A.1.a.3, name=Oral Expression, categories=['Abilities'], {'competencyText': 'The ability to communicate information and ideas in speaking so others will understand.'}),\n",
       " Competency(identifier=1.A.1.a.4, name=Written Expression, categories=['Abilities'], {'competencyText': 'The ability to communicate information and ideas in writing so others will understand.'}),\n",
       " Competency(identifier=1.A.1.b.1, name=Fluency of Ideas, categories=['Abilities'], {'competencyText': 'The ability to come up with a number of ideas about a topic (the number of ideas is important, not their quality, correctness, or creativity).'}),\n",
       " Competency(identifier=1.A.1.b.2, name=Originality, categories=['Abilities'], {'competencyText': 'The ability to come up with unusual or clever ideas about a given topic or situation, or to develop creative ways to solve a problem.'}),\n",
       " Competency(identifier=1.A.1.b.3, name=Problem Sensitivity, categories=['Abilities'], {'competencyText': 'The ability to tell when something is wrong or is likely to go wrong. It does not involve solving the problem, only recognizing there is a problem.'}),\n",
       " Competency(identifier=1.A.1.b.4, name=Deductive Reasoning, categories=['Abilities'], {'competencyText': 'The ability to apply general rules to specific problems to produce answers that make sense.'}),\n",
       " Competency(identifier=1.A.1.b.5, name=Inductive Reasoning, categories=['Abilities'], {'competencyText': 'The ability to combine pieces of information to form general rules or conclusions (includes finding a relationship among seemingly unrelated events).'}),\n",
       " Competency(identifier=1.A.1.b.6, name=Information Ordering, categories=['Abilities'], {'competencyText': 'The ability to arrange things or actions in a certain order or pattern according to a specific rule or set of rules (e.g., patterns of numbers, letters, words, pictures, mathematical operations).'}),\n",
       " Competency(identifier=1.A.1.b.7, name=Category Flexibility, categories=['Abilities'], {'competencyText': 'The ability to generate or use different sets of rules for combining or grouping things in different ways.'}),\n",
       " Competency(identifier=1.A.1.c.1, name=Mathematical Reasoning, categories=['Abilities'], {'competencyText': 'The ability to choose the right mathematical methods or formulas to solve a problem.'}),\n",
       " Competency(identifier=1.A.1.d.1, name=Memorization, categories=['Abilities'], {'competencyText': 'The ability to remember information such as words, numbers, pictures, and procedures.'}),\n",
       " Competency(identifier=1.A.1.e.1, name=Speed of Closure, categories=['Abilities'], {'competencyText': 'The ability to quickly make sense of, combine, and organize information into meaningful patterns.'}),\n",
       " Competency(identifier=1.A.1.e.2, name=Flexibility of Closure, categories=['Abilities'], {'competencyText': 'The ability to identify or detect a known pattern (a figure, object, word, or sound) that is hidden in other distracting material.'}),\n",
       " Competency(identifier=1.A.1.e.3, name=Perceptual Speed, categories=['Abilities'], {'competencyText': 'The ability to quickly and accurately compare similarities and differences among sets of letters, numbers, objects, pictures, or patterns. The things to be compared may be presented at the same time or one after the other. This ability also includes comparing a presented object with a remembered object.'}),\n",
       " Competency(identifier=1.A.2.a.1, name=Arm-Hand Steadiness, categories=['Abilities'], {'competencyText': 'The ability to keep your hand and arm steady while moving your arm or while holding your arm and hand in one position.'}),\n",
       " Competency(identifier=1.A.2.a.3, name=Finger Dexterity, categories=['Abilities'], {'competencyText': 'The ability to make precisely coordinated movements of the fingers of one or both hands to grasp, manipulate, or assemble very small objects.'}),\n",
       " Competency(identifier=1.A.4.a.1, name=Near Vision, categories=['Abilities'], {'competencyText': 'The ability to see details at close range (within a few feet of the observer).'}),\n",
       " Competency(identifier=1.A.4.b.4, name=Speech Recognition, categories=['Abilities'], {'competencyText': 'The ability to identify and understand the speech of another person.'}),\n",
       " Competency(identifier=1.A.4.b.5, name=Speech Clarity, categories=['Abilities'], {'competencyText': 'The ability to speak clearly so others can understand you.'}),\n",
       " Competency(identifier=2.A.1.a, name=Reading Comprehension, categories=['Skills'], {'competencyText': 'Understanding written sentences and paragraphs in work related documents.'}),\n",
       " Competency(identifier=2.A.1.b, name=Active Listening, categories=['Skills'], {'competencyText': 'Giving full attention to what other people are saying, taking time to understand the points being made, asking questions as appropriate, and not interrupting at inappropriate times.'}),\n",
       " Competency(identifier=2.A.1.c, name=Writing, categories=['Skills'], {'competencyText': 'Communicating effectively in writing as appropriate for the needs of the audience.'}),\n",
       " Competency(identifier=2.A.1.d, name=Speaking, categories=['Skills'], {'competencyText': 'Talking to others to convey information effectively.'}),\n",
       " Competency(identifier=2.A.1.f, name=Science, categories=['Skills'], {'competencyText': 'Using scientific rules and methods to solve problems.'}),\n",
       " Competency(identifier=2.A.2.a, name=Critical Thinking, categories=['Skills'], {'competencyText': 'Using logic and reasoning to identify the strengths and weaknesses of alternative solutions, conclusions or approaches to problems.'}),\n",
       " Competency(identifier=2.A.2.b, name=Active Learning, categories=['Skills'], {'competencyText': 'Understanding the implications of new information for both current and future problem-solving and decision-making.'}),\n",
       " Competency(identifier=2.A.2.c, name=Learning Strategies, categories=['Skills'], {'competencyText': 'Selecting and using training/instructional methods and procedures appropriate for the situation when learning or teaching new things.'}),\n",
       " Competency(identifier=2.A.2.d, name=Monitoring, categories=['Skills'], {'competencyText': 'Monitoring/Assessing performance of yourself, other individuals, or organizations to make improvements or take corrective action.'}),\n",
       " Competency(identifier=2.B.1.a, name=Social Perceptiveness, categories=['Skills'], {'competencyText': \"Being aware of others' reactions and understanding why they react as they do.\"}),\n",
       " Competency(identifier=2.B.1.b, name=Coordination, categories=['Skills'], {'competencyText': \"Adjusting actions in relation to others' actions.\"}),\n",
       " Competency(identifier=2.B.1.c, name=Persuasion, categories=['Skills'], {'competencyText': 'Persuading others to change their minds or behavior.'}),\n",
       " Competency(identifier=2.B.1.d, name=Negotiation, categories=['Skills'], {'competencyText': 'Bringing others together and trying to reconcile differences.'}),\n",
       " Competency(identifier=2.B.1.e, name=Instructing, categories=['Skills'], {'competencyText': 'Teaching others how to do something.'}),\n",
       " Competency(identifier=2.B.1.f, name=Service Orientation, categories=['Skills'], {'competencyText': 'Actively looking for ways to help people.'}),\n",
       " Competency(identifier=2.B.2.i, name=Complex Problem Solving, categories=['Skills'], {'competencyText': 'Identifying complex problems and reviewing related information to develop and evaluate options and implement solutions.'}),\n",
       " Competency(identifier=2.B.3.a, name=Operations Analysis, categories=['Skills'], {'competencyText': 'Analyzing needs and product requirements to create a design.'}),\n",
       " Competency(identifier=2.B.4.e, name=Judgment and Decision Making, categories=['Skills'], {'competencyText': 'Considering the relative costs and benefits of potential actions to choose the most appropriate one.'}),\n",
       " Competency(identifier=2.B.4.g, name=Systems Analysis, categories=['Skills'], {'competencyText': 'Determining how a system should work and how changes in conditions, operations, and the environment will affect outcomes.'}),\n",
       " Competency(identifier=2.B.4.h, name=Systems Evaluation, categories=['Skills'], {'competencyText': 'Identifying measures or indicators of system performance and the actions needed to improve or correct performance, relative to the goals of the system.'}),\n",
       " Competency(identifier=2.B.5.a, name=Time Management, categories=['Skills'], {'competencyText': \"Managing one's own time and the time of others.\"}),\n",
       " Competency(identifier=2.B.5.d, name=Management of Personnel Resources, categories=['Skills'], {'competencyText': 'Motivating, developing, and directing people as they work, identifying the best people for the job.'}),\n",
       " Competency(identifier=2.C.1.b, name=Clerical, categories=['Knowledge'], {'competencyText': 'Knowledge of administrative and clerical procedures and systems such as word processing, managing files and records, stenography and transcription, designing forms, and other office procedures and terminology.'}),\n",
       " Competency(identifier=2.C.1.e, name=Customer and Personal Service, categories=['Knowledge'], {'competencyText': 'Knowledge of principles and processes for providing customer and personal services. This includes customer needs assessment, meeting quality standards for services, and evaluation of customer satisfaction.'}),\n",
       " Competency(identifier=2.C.3.a, name=Computers and Electronics, categories=['Knowledge'], {'competencyText': 'Knowledge of circuit boards, processors, chips, electronic equipment, and computer hardware and software, including applications and programming.'}),\n",
       " Competency(identifier=2.C.4.a, name=Mathematics, categories=['Knowledge'], {'competencyText': 'Knowledge of arithmetic, algebra, geometry, calculus, statistics, and their applications.'}),\n",
       " Competency(identifier=2.C.4.c, name=Chemistry, categories=['Knowledge'], {'competencyText': 'Knowledge of the chemical composition, structure, and properties of substances and of the chemical processes and transformations that they undergo. This includes uses of chemicals and their interactions, danger signs, production techniques, and disposal methods.'}),\n",
       " Competency(identifier=2.C.4.d, name=Biology, categories=['Knowledge'], {'competencyText': 'Knowledge of plant and animal organisms, their tissues, cells, functions, interdependencies, and interactions with each other and the environment.'}),\n",
       " Competency(identifier=2.C.4.e, name=Psychology, categories=['Knowledge'], {'competencyText': 'Knowledge of human behavior and performance; individual differences in ability, personality, and interests; learning and motivation; psychological research methods; and the assessment and treatment of behavioral and affective disorders.'}),\n",
       " Competency(identifier=2.C.4.f, name=Sociology and Anthropology, categories=['Knowledge'], {'competencyText': 'Knowledge of group behavior and dynamics, societal trends and influences, human migrations, ethnicity, cultures and their history and origins.'}),\n",
       " Competency(identifier=2.C.5.a, name=Medicine and Dentistry, categories=['Knowledge'], {'competencyText': 'Knowledge of the information and techniques needed to diagnose and treat human injuries, diseases, and deformities. This includes symptoms, treatment alternatives, drug properties and interactions, and preventive health-care measures.'}),\n",
       " Competency(identifier=2.C.5.b, name=Therapy and Counseling, categories=['Knowledge'], {'competencyText': 'Knowledge of principles, methods, and procedures for diagnosis, treatment, and rehabilitation of physical and mental dysfunctions, and for career counseling and guidance.'}),\n",
       " Competency(identifier=2.C.6, name=Education and Training, categories=['Knowledge'], {'competencyText': 'Knowledge of principles and methods for curriculum and training design, teaching and instruction for individuals and groups, and the measurement of training effects.'}),\n",
       " Competency(identifier=2.C.7.a, name=English Language, categories=['Knowledge'], {'competencyText': 'Knowledge of the structure and content of the English language including the meaning and spelling of words, rules of composition, and grammar.'}),\n",
       " Competency(identifier=2.C.7.e, name=Philosophy and Theology, categories=['Knowledge'], {'competencyText': 'Knowledge of different philosophical systems and religions. This includes their basic principles, values, ethics, ways of thinking, customs, practices, and their impact on human culture.'}),\n",
       " Competency(identifier=2.C.8.a, name=Public Safety and Security, categories=['Knowledge'], {'competencyText': 'Knowledge of relevant equipment, policies, procedures, and strategies to promote effective local, state, or national security operations for the protection of people, data, property, and institutions.'}),\n",
       " Competency(identifier=2.C.8.b, name=Law and Government, categories=['Knowledge'], {'competencyText': 'Knowledge of laws, legal codes, court procedures, precedents, government regulations, executive orders, agency rules, and the democratic political process.'}),\n",
       " Competency(identifier=2.C.9.b, name=Communications and Media, categories=['Knowledge'], {'competencyText': 'Knowledge of media production, communication, and dissemination techniques and methods. This includes alternative ways to inform and entertain via written, oral, and visual media.'}),\n",
       " Competency(identifier=41103901-Microhematocrit centrifuges, name=Microhematocrit centrifuges, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=41104102-Lancets, name=Lancets, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=41104104-Tourniquets, name=Tourniquets, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=41104107-Evacuated blood collection tubes, name=Evacuated blood collection tubes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=41104118-Specimen collection containers, name=Specimen collection containers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=41104403-Tissue culture incubators, name=Tissue culture incubators, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=41111709-Binocular light compound microscopes, name=Binocular light compound microscopes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=41115807-Hemoglobin analyzers, name=Hemoglobin analyzers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=41116138-Urinalysis test strips, name=Urinalysis test strips, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=41116201-Glucometers, name=Glucometers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42131606-Protective face shields, name=Protective face shields, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42131612-Protective gowns, name=Protective gowns, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42132203-Medical examination protective gloves, name=Medical examination protective gloves, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42141807-Transcutaneous electric nerve stimulation TENS equipment, name=Transcutaneous electric nerve stimulation TENS equipment, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42142404-Nasal suctioning equipment, name=Nasal suctioning equipment, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42142404-Oral suctioning equipment, name=Oral suctioning equipment, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42142404-Tracheal suctioning equipment, name=Tracheal suctioning equipment, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42142509-Epidural catheters, name=Epidural catheters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42142532-Pericardiocentesis kits, name=Pericardiocentesis kits, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42142537-Thoracentesis kits, name=Thoracentesis kits, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42142609-Hypodermic syringes, name=Hypodermic syringes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42142616-Blood drawing syringes, name=Blood drawing syringes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42142715-Urinary catheters, name=Urinary catheters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42144102-Chest tubes, name=Chest tubes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42171608-Head immobilization devices, name=Head immobilization devices, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42171613-Spinal immobilization equipment, name=Spinal immobilization equipment, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42172101-Automated external defibrillators AED, name=Automated external defibrillators AED, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42172102-Cardiopulmonary resuscitation CPR face shields, name=Cardiopulmonary resuscitation CPR face shields, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42181602-Electronic blood pressure monitors, name=Electronic blood pressure monitors, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42181602-Pediatric blood pressure cuffs, name=Pediatric blood pressure cuffs, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42181701-Electrocardiography EKG machines, name=Electrocardiography EKG machines, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42181701-Portable electrocardiography EKG machines, name=Portable electrocardiography EKG machines, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42181713-Holter monitors, name=Holter monitors, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42181801-Pulse oximeters, name=Pulse oximeters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42181902-Intracranial pressure monitors, name=Intracranial pressure monitors, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42181903-Cardiac monitors, name=Cardiac monitors, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42181903-Hemodynamic monitors, name=Hemodynamic monitors, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42182005-Ophthalmoscopes, name=Ophthalmoscopes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42182005-Otoscopes, name=Otoscopes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42182013-Vaginal exam specula, name=Vaginal exam specula, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42182103-Mechanical stethoscopes, name=Mechanical stethoscopes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42182201-Digital medical thermometers, name=Digital medical thermometers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42182302-Reflex hammers, name=Reflex hammers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42182412-Diagnostic tuning forks, name=Diagnostic tuning forks, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42182416-Tympanometers, name=Tympanometers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42182805-Medical scales, name=Medical scales, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42183001-Snellen eye charts, name=Snellen eye charts, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42183001-Visual acuity testing cards, name=Visual acuity testing cards, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42192401-Crash carts, name=Crash carts, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42201701-Doppler ultrasound equipment, name=Doppler ultrasound equipment, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42203402-Angiocaths, name=Angiocaths, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42203402-Pulmonary artery catheters, name=Pulmonary artery catheters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42203501-Pacemaker analyzers, name=Pacemaker analyzers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42203501-Transcutaneous pacemakers, name=Transcutaneous pacemakers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42211502-Crutches, name=Crutches, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42221501-Arterial line catheters, name=Arterial line catheters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42221503-Multiple lumen central line catheters, name=Multiple lumen central line catheters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42221506-Umbilical catheters, name=Umbilical catheters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42221509-Intravenous IV administration sets, name=Intravenous IV administration sets, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42221509-Intravenous IV cutdown trays, name=Intravenous IV cutdown trays, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42222001-Intravenous IV infusion pumps, name=Intravenous IV infusion pumps, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42231701-Nasogastric tubes, name=Nasogastric tubes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42241507-Orthopedic splinting equipment, name=Orthopedic splinting equipment, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42241705-Lower extremity braces, name=Lower extremity braces, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42241707-Walking braces, name=Walking braces, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42241801-Upper extremity braces, name=Upper extremity braces, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42241802-Back braces, name=Back braces, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42241803-Neck braces, name=Neck braces, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42242103-Halo traction equipment, name=Halo traction equipment, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271501-Apnea monitors, name=Apnea monitors, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271502-Arterial blood gas monitoring equipment, name=Arterial blood gas monitoring equipment, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271602-Incentive spirometers, name=Incentive spirometers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271604-Peak flowmeters, name=Peak flowmeters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271702-Oxygen concentrators, name=Oxygen concentrators, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271707-Oxygen flowmeters, name=Oxygen flowmeters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271708-Oxygen delivery masks, name=Oxygen delivery masks, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271710-Nasal catheters, name=Nasal catheters, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271802-Handheld nebulizers, name=Handheld nebulizers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271903-Endotracheal ET tubes, name=Endotracheal ET tubes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42271910-Tracheotomy sets, name=Tracheotomy sets, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42272001-Fiberoptic laryngoscopes, name=Fiberoptic laryngoscopes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42272017-Intubation sets, name=Intubation sets, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42272201-Mechanical intermittent positive pressure ventilators, name=Mechanical intermittent positive pressure ventilators, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42272203-Bilevel positive airway pressure BiPAP ventilators, name=Bilevel positive airway pressure BiPAP ventilators, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42291613-Surgical scalpels, name=Surgical scalpels, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42291614-Straight surgical scissors, name=Straight surgical scissors, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42291704-Biopsy punches, name=Biopsy punches, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42291802-Mosquito hemostats, name=Mosquito hemostats, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42294702-Intra-aortic balloon pumps IABP, name=Intra-aortic balloon pumps IABP, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42294717-Ventricular assist devices VAD, name=Ventricular assist devices VAD, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42294802-Fiberoptic endoscopes, name=Fiberoptic endoscopes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42294802-Flexible sigmoidoscopes, name=Flexible sigmoidoscopes, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42295104-Electrosurgical cauterization machines, name=Electrosurgical cauterization machines, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42295119-Argon lasers, name=Argon lasers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42295119-Carbon dioxide CO2 lasers, name=Carbon dioxide CO2 lasers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42295119-Pulsed dye lasers, name=Pulsed dye lasers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42295122-Pneumatic tourniquets, name=Pneumatic tourniquets, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42312008-Surgical staple removers, name=Surgical staple removers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42312010-Skin staplers, name=Skin staplers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=42312202-Suturing kits, name=Suturing kits, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=43191507-Multi-line telephone systems, name=Multi-line telephone systems, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=43211503-Laptop computers, name=Laptop computers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=43211504-Personal digital assistants PDA, name=Personal digital assistants PDA, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=43211508-Personal computers, name=Personal computers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=43211509-Tablet computers, name=Tablet computers, categories=['Tools', 'O*NET T2'], {}),\n",
       " Competency(identifier=43231507-Microsoft SharePoint, name=Microsoft SharePoint, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43231513-Microsoft Office, name=Microsoft Office, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232104-Microsoft Word, name=Microsoft Word, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232106-Microsoft PowerPoint, name=Microsoft PowerPoint, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232110-Microsoft Excel, name=Microsoft Excel, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232306-Microsoft Access, name=Microsoft Access, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-Allscripts Professional EHR, name=Allscripts Professional EHR, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-Amkai AmkaiCharts, name=Amkai AmkaiCharts, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-Bizmatics PrognoCIS EMR, name=Bizmatics PrognoCIS EMR, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-Cerner Millennium, name=Cerner Millennium, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-GE Healthcare Centricity EMR, name=GE Healthcare Centricity EMR, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-Healthcare common procedure coding system HCPCS, name=Healthcare common procedure coding system HCPCS, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-MEDITECH software, name=MEDITECH software, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-Medical condition coding software, name=Medical condition coding software, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-Medical procedure coding software, name=Medical procedure coding software, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-Medscribbler Enterprise, name=Medscribbler Enterprise, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-MicroFour PracticeStudio.NET EMR, name=MicroFour PracticeStudio.NET EMR, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-NextGen Healthcare Information Systems EMR, name=NextGen Healthcare Information Systems EMR, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-PCC Pediatric Partner, name=PCC Pediatric Partner, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-SOAPware EMR, name=SOAPware EMR, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-StatCom Patient Flow Logistics Enterprise Suite, name=StatCom Patient Flow Logistics Enterprise Suite, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-SynaMed EMR, name=SynaMed EMR, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-Texas Medical Software SpringCharts EMR, name=Texas Medical Software SpringCharts EMR, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-e-MDs software, name=e-MDs software, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232610-eClinicalWorks, name=eClinicalWorks, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232705-Microsoft Internet Explorer, name=Microsoft Internet Explorer, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43232705-Web browser software, name=Web browser software, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=43233501-Microsoft Outlook, name=Microsoft Outlook, categories=['Technology', 'O*NET T2'], {}),\n",
       " Competency(identifier=46181804-Safety goggles, name=Safety goggles, categories=['Tools', 'O*NET T2'], {})}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nurse_practitioners = onet.filter_by(lambda edge: 'Nurse Practitioners' in edge.occupation.name)\n",
    "nurse_practitioners.competencies"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That's a big list of competencies. Each competency has a list of categories, so let's get all of the different possible categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Abilities', 'Knowledge', 'O*NET T2', 'Skills', 'Technology', 'Tools'}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(cat for competency in nurse_practitioners.competencies for cat in competency.categories)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So we can ask questions like: what type of Knowledge do nurse practitioners need?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Competency(identifier=2.C.7.a, name=English Language, categories=['Knowledge'], {'competencyText': 'Knowledge of the structure and content of the English language including the meaning and spelling of words, rules of composition, and grammar.'}),\n",
       " Competency(identifier=2.C.1.b, name=Clerical, categories=['Knowledge'], {'competencyText': 'Knowledge of administrative and clerical procedures and systems such as word processing, managing files and records, stenography and transcription, designing forms, and other office procedures and terminology.'}),\n",
       " Competency(identifier=2.C.4.a, name=Mathematics, categories=['Knowledge'], {'competencyText': 'Knowledge of arithmetic, algebra, geometry, calculus, statistics, and their applications.'}),\n",
       " Competency(identifier=2.C.8.b, name=Law and Government, categories=['Knowledge'], {'competencyText': 'Knowledge of laws, legal codes, court procedures, precedents, government regulations, executive orders, agency rules, and the democratic political process.'}),\n",
       " Competency(identifier=2.C.6, name=Education and Training, categories=['Knowledge'], {'competencyText': 'Knowledge of principles and methods for curriculum and training design, teaching and instruction for individuals and groups, and the measurement of training effects.'}),\n",
       " Competency(identifier=2.C.4.f, name=Sociology and Anthropology, categories=['Knowledge'], {'competencyText': 'Knowledge of group behavior and dynamics, societal trends and influences, human migrations, ethnicity, cultures and their history and origins.'}),\n",
       " Competency(identifier=2.C.7.e, name=Philosophy and Theology, categories=['Knowledge'], {'competencyText': 'Knowledge of different philosophical systems and religions. This includes their basic principles, values, ethics, ways of thinking, customs, practices, and their impact on human culture.'}),\n",
       " Competency(identifier=2.C.4.d, name=Biology, categories=['Knowledge'], {'competencyText': 'Knowledge of plant and animal organisms, their tissues, cells, functions, interdependencies, and interactions with each other and the environment.'}),\n",
       " Competency(identifier=2.C.4.c, name=Chemistry, categories=['Knowledge'], {'competencyText': 'Knowledge of the chemical composition, structure, and properties of substances and of the chemical processes and transformations that they undergo. This includes uses of chemicals and their interactions, danger signs, production techniques, and disposal methods.'}),\n",
       " Competency(identifier=2.C.8.a, name=Public Safety and Security, categories=['Knowledge'], {'competencyText': 'Knowledge of relevant equipment, policies, procedures, and strategies to promote effective local, state, or national security operations for the protection of people, data, property, and institutions.'}),\n",
       " Competency(identifier=2.C.5.a, name=Medicine and Dentistry, categories=['Knowledge'], {'competencyText': 'Knowledge of the information and techniques needed to diagnose and treat human injuries, diseases, and deformities. This includes symptoms, treatment alternatives, drug properties and interactions, and preventive health-care measures.'}),\n",
       " Competency(identifier=2.C.5.b, name=Therapy and Counseling, categories=['Knowledge'], {'competencyText': 'Knowledge of principles, methods, and procedures for diagnosis, treatment, and rehabilitation of physical and mental dysfunctions, and for career counseling and guidance.'}),\n",
       " Competency(identifier=2.C.4.e, name=Psychology, categories=['Knowledge'], {'competencyText': 'Knowledge of human behavior and performance; individual differences in ability, personality, and interests; learning and motivation; psychological research methods; and the assessment and treatment of behavioral and affective disorders.'}),\n",
       " Competency(identifier=2.C.9.b, name=Communications and Media, categories=['Knowledge'], {'competencyText': 'Knowledge of media production, communication, and dissemination techniques and methods. This includes alternative ways to inform and entertain via written, oral, and visual media.'}),\n",
       " Competency(identifier=2.C.1.e, name=Customer and Personal Service, categories=['Knowledge'], {'competencyText': 'Knowledge of principles and processes for providing customer and personal services. This includes customer needs assessment, meeting quality standards for services, and evaluation of customer satisfaction.'}),\n",
       " Competency(identifier=2.C.3.a, name=Computers and Electronics, categories=['Knowledge'], {'competencyText': 'Knowledge of circuit boards, processors, chips, electronic equipment, and computer hardware and software, including applications and programming.'})]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[competency for competency in nurse_practitioners.competencies if 'Knowledge' in competency.categories]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are many other questions we can ask of just an ontology, but the real value will come from combining the knowledge contained in the ontology with larger unstructured datasets. In this next section we will explore how Skills-ML helps the user deal with such datasets as job postings, profiles, or course descriptions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Spirit of Skills-ML\n",
    "\n",
    "- **Dataflow Programming**: `Skills-ML`'s design philosophy builds on dataflow programming or so called data streaming to process very large datasets (larger than RAM; potentially infinite). \n",
    "- **One-pass algorithm**: Data points are processed one at a time.\n",
    "- **Lazy evaluation**: an evaluation strategy which delays the evaluation of an expression until its value is needed. In `Skills-ML`, most of the classes and functions here incorporates the concept of `Iterable` or `Generator`. We build the expression first and evaluate later. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Creating Dataset\n",
    "\n",
    "Before we do anything with the context, we need dataset. `Skills-ML` makes use of schema.org’s JobPosting standard. As it has been in use for a long time, some open sources are already using this standard, which is easy to import. Other job posting data sources are converted into the schema.org Schema and all work on job postings is done using this standard schema. \n",
    "\n",
    "In `Skills-ML`, `job_postings` module has all the functionalities to create the data we need for later usage. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Common Schema\n",
    "\n",
    "We have an useful function to help create the data generator from s3. \n",
    "\n",
    "- `JobPostingCollectionFromS3`: Stream job posting from s3.\n",
    "- `JobPostingCollectionSample`: Stream a finite number of job postings stored within the library.\n",
    "\n",
    "However, we are not restrcted to just JobPosting data. One can easily create whatever data generator such as `ProfileGenerator` or `CourseGenerator`. \n",
    "\n",
    "For example, we want to use the Vrigina Dataset which is an open data set of job postings. We just have to create a job posting generator with some transformation. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.job_postings.raw.virginia import VirginiaTransformer\n",
    "from urllib.request import urlopen\n",
    "import json\n",
    "\n",
    "va_url = \"http://opendata.cs.vt.edu/dataset/ab0abac3-2293-4c9d-8d80-22d450254389/resource/074f7e44-9275-4bba-874e-4795e8f6830c/download/openjobs-jobpostings.may-2016.json\"\n",
    "\n",
    "class VAJobposting(object):\n",
    "    def __init__(self, uri):\n",
    "        self.uri = uri\n",
    "    \n",
    "    def __iter__(self):\n",
    "        request = urlopen(self.uri)\n",
    "        for line in request.readlines():\n",
    "            raw = json.loads(line)\n",
    "            yield VirginiaTransformer(partner_id=\"va\")._transform(raw)\n",
    "            \n",
    "\n",
    "jobpostings_va = VAJobposting(va_url)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "40098\n"
     ]
    }
   ],
   "source": [
    "print(len(list(jobpostings_va)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Filtering\n",
    "\n",
    "To create a good dataset, we might want to have some criteria for choosing the proper job posting based on the task we want to perform, like job postings that have the label information, job postings that belong to certain occupation, or job postings that have rich enough information in the description field.\n",
    "\n",
    "- `JobPostingFilterer`: Filter common schema job postings through a number of filtering functions.\n",
    "\n",
    "This function also follows lazy evaluation strategy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.job_postings.filtering import JobPostingFilterer\n",
    "\n",
    "def is_tech_jobs(job):\n",
    "    if job['onet_soc_code'][:2] in ['15', '17', '19']:\n",
    "        return True\n",
    "    else:\n",
    "        return False\n",
    "    \n",
    "tech_jobs = JobPostingFilterer(\n",
    "    job_posting_generator=VAJobposting(va_url), \n",
    "    filter_funcs=[is_tech_jobs]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.ontologies.onet import majorgroupname\n",
    "from collections import Counter\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "import seaborn as sns\n",
    "sns.set(style=\"darkgrid\", font_scale=2)\n",
    "%matplotlib inline\n",
    "\n",
    "# major group distribution plotting function\n",
    "\n",
    "def plot_major_group_distribution(job_postings):\n",
    "    c = Counter()\n",
    "    for job in job_postings:\n",
    "        c.update([job['onet_soc_code'][:2]])\n",
    "    s = pd.Series(c).sort_index()\n",
    "    s.index = s.index.map(majorgroupname)\n",
    "    ax = s.plot.bar(figsize=(20,10),rot=90)\n",
    "    ax.set_xlabel('soc_major_group')\n",
    "    ax.set_ylabel('number of job posting')\n",
    "    ax.set_title(f\"total number: {s.sum()}\")\n",
    "    return s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Computer and Mathematical             5065\n",
       "Architecture and Engineering          1937\n",
       "Life, Physical, and Social Science     192\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_major_group_distribution(tech_jobs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What if we want to make sure that all the job postings have ONet SOC Code and it's not unknown(first 2 digit 99)? We can define filter functions like these which can be either generic function or lambda function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def filter_onet_soc_code(job):\n",
    "    if job['onet_soc_code'] and job['onet_soc_code'][:2] != '99':\n",
    "        return True\n",
    "    else:\n",
    "        return False\n",
    "\n",
    "has_soc = lambda x: x['onet_soc_code']\n",
    "not_unknown_soc = lambda x: x['onet_soc_code'][:2] != '99'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "jobpostings_filtered = JobPostingFilterer(\n",
    "    job_posting_generator=VAJobposting(va_url), \n",
    "    filter_funcs=[has_soc, not_unknown_soc]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Management                                        6506\n",
       "Business and Financial Operations                 3867\n",
       "Computer and Mathematical                         5065\n",
       "Architecture and Engineering                      1937\n",
       "Life, Physical, and Social Science                 192\n",
       "Community and Social Service                       282\n",
       "Legal                                               94\n",
       "Education, Training, and Library                   679\n",
       "Arts, Design, Entertainment, Sports, and Media     598\n",
       "Healthcare Practitioners and Technical            3447\n",
       "Healthcare Support                                 494\n",
       "Protective Service                                 484\n",
       "Food Preparation and Serving Related               792\n",
       "Building and Grounds Cleaning and Maintenance      189\n",
       "Personal Care and Service                           97\n",
       "Sales and Related                                 1415\n",
       "Office and Administrative Support                 2580\n",
       "Construction and Extraction                        196\n",
       "Installation, Maintenance, and Repair              832\n",
       "Production                                         442\n",
       "Transportation and Material Moving                2722\n",
       "Military Specific                                    2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_major_group_distribution(jobpostings_filtered)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Sampling\n",
    "\n",
    "Even though we have a lot of data, most of time we don't need all of them to do the analysis. Or we can't even fit all the data into memory to do the analysis. What we need more importantly is a suitable sampled dataset.\n",
    "\n",
    "- `JobSampler`: Sample job posting by (weighted) reservoir sampling. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Sampling from Streaming Data - Reservoir Sampling\n",
    "\n",
    "- \"Say you have a stream of items of large and unknown length that we can only iterate over once.\"\n",
    "- It's memeory efficient and just one iteration\n",
    "- There is a great overview of reservoir sampling in https://gregable.com/2007/10/reservoir-sampling.html. \n",
    "\n",
    "Let's say the original job postings dataset are too much for my Mac Yosemite to do any analysis and I want only 1000 job postings but still preserve the statistical characteristics of the original dataset. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.job_postings.sample import JobSampler\n",
    "\n",
    "sampler = JobSampler(\n",
    "    job_posting_generator=jobpostings_filtered, \n",
    "    k=1000,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Management                                        191\n",
       "Business and Financial Operations                 126\n",
       "Computer and Mathematical                         134\n",
       "Architecture and Engineering                       65\n",
       "Life, Physical, and Social Science                  5\n",
       "Community and Social Service                       11\n",
       "Legal                                               1\n",
       "Education, Training, and Library                   25\n",
       "Arts, Design, Entertainment, Sports, and Media     15\n",
       "Healthcare Practitioners and Technical            106\n",
       "Healthcare Support                                 20\n",
       "Protective Service                                 16\n",
       "Food Preparation and Serving Related               27\n",
       "Building and Grounds Cleaning and Maintenance       5\n",
       "Personal Care and Service                           1\n",
       "Sales and Related                                  51\n",
       "Office and Administrative Support                  75\n",
       "Construction and Extraction                         6\n",
       "Installation, Maintenance, and Repair              19\n",
       "Production                                         12\n",
       "Transportation and Material Moving                 89\n",
       "dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_major_group_distribution(sampler)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Something wrong happened! We are missing `Military Occupations`! Because military job postings are extremely rare in the original dataset, simple ramdom sampling might result in lack of classes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Weighted Reservoir Sampling\n",
    "\n",
    "- How would you sample from a weighted distribution where each element has a given weight associated with it in the stream?\n",
    "\n",
    "- For certain task, we need some curated sample. For example, if we want to build a occupation classifier, we want similar amounts of job posting for each occupation. \n",
    "\n",
    "Now we want to have a more uniform distributed sample across all major groups. \n",
    "Here we need to provide a weight dictionary in the `JobSampler`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "c = Counter()\n",
    "for job in jobpostings_filtered:\n",
    "    c.update([job['onet_soc_code'][:2]])\n",
    "weights = dict()\n",
    "for key, value in c.items():\n",
    "    weights[key] = max(c.values()) / value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'11': 1.0,\n",
       " '15': 1.2845014807502468,\n",
       " '17': 3.3588022715539494,\n",
       " '29': 1.8874383521903104,\n",
       " '41': 4.597879858657244,\n",
       " '43': 2.521705426356589,\n",
       " '13': 1.6824411688647531,\n",
       " '49': 7.819711538461538,\n",
       " '33': 13.442148760330578,\n",
       " '27': 10.879598662207357,\n",
       " '47': 33.19387755102041,\n",
       " '51': 14.71945701357466,\n",
       " '35': 8.214646464646465,\n",
       " '25': 9.581737849779087,\n",
       " '31': 13.17004048582996,\n",
       " '19': 33.885416666666664,\n",
       " '21': 23.070921985815602,\n",
       " '37': 34.423280423280424,\n",
       " '53': 2.3901542983100663,\n",
       " '39': 67.0721649484536,\n",
       " '23': 69.2127659574468,\n",
       " '55': 3253.0}"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "sampler = JobSampler(job_posting_generator=jobpostings_filtered, k=1000, key=lambda x: x['onet_soc_code'][:2], weights=weights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Management                                        54\n",
       "Business and Financial Operations                 47\n",
       "Computer and Mathematical                         58\n",
       "Architecture and Engineering                      47\n",
       "Life, Physical, and Social Science                39\n",
       "Community and Social Service                      53\n",
       "Legal                                             43\n",
       "Education, Training, and Library                  47\n",
       "Arts, Design, Entertainment, Sports, and Media    54\n",
       "Healthcare Practitioners and Technical            43\n",
       "Healthcare Support                                46\n",
       "Protective Service                                48\n",
       "Food Preparation and Serving Related              41\n",
       "Building and Grounds Cleaning and Maintenance     49\n",
       "Personal Care and Service                         36\n",
       "Sales and Related                                 48\n",
       "Office and Administrative Support                 48\n",
       "Construction and Extraction                       62\n",
       "Installation, Maintenance, and Repair             44\n",
       "Production                                        44\n",
       "Transportation and Material Moving                47\n",
       "Military Specific                                  2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_major_group_distribution(sampler)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Skill Extraction\n",
    "\n",
    "A common task is extracting competencies from unstructured text. Sometimes this is ontology-based (finding concepts from a known ontology in text), but this is not necessarily true. Skills-ML unites these with a common interface in the SkillExtractor class. The common interface is that every SkillExtractor needs to be able to take in a collection of documents, and yield what we call `CandidateSkill` objects.\n",
    "\n",
    "## What Is a CandidateSkill?\n",
    "\n",
    "A CandidateSkill is a possible occurrence of a skill/competency in context in some document. It consists of the following fields:\n",
    "\n",
    "`skill_name` - The text version of the skill as it appears in the document\n",
    "\n",
    "`matched_skill_identifier` - A reference to the skill in some ontology. This may be empty, if no ontology was used to search for skills.\n",
    "\n",
    "`context` - The text surrounding the skill in the document. The goal is for a human labeler to be able to use this to determine whether or not the occurrence represents a true skill. How much context is included is up to the algorithm.\n",
    "\n",
    "`start_index` - The start index of the skill occurrence within the document string.\n",
    "\n",
    "`confidence` - The confidence level the algorithm has in this candidate skill being a true occurrence of a skill. This may be empty, if the algorithm has now way of producing a confidence value.\n",
    "\n",
    "`document_id` - A unique identifier for the source document.\n",
    "\n",
    "`document_type` - The type of document (examples: Job Posting, Profile, Course Description)\n",
    "\n",
    "`source_object` - The entire source document.\n",
    "\n",
    "`skill_extractor_name` - The name of the skill extractor algorithm. Every SkillExtractor subclass defines a name property that is used to give processes downstream context about how their output data was produced.\n",
    "\n",
    "The idea behind the CandidateSkill object is to serve as a common interface between SkillExtractor objects, automatic evaluation methods, and manual evaluation methods.\n",
    "\n",
    "- A labeling interface might intake CandidateSkill objects for humans to say yes/no to.\n",
    "- Another type of labeling interface might involve the export of CandidateSkill objects based on what a human highlighted in the interface when shown the entire document\n",
    "- Unsupervised evaluation metrics may take in one set of CandidateSkills to produce simple descriptive metrics\n",
    "- Supervised evaluation metrics may take in one set of CandidateSkills from a SkillExtractor and another set of CandidateSkills from a human labeling interface and use the latter to evaluate the former\n",
    "\n",
    "We'll talk about some of these use cases in more detail later. But for now, let's start with a simple example that uses NLP rules and isn't tied to an ontology. Let's define a method for extracting skills as 'all noun phrases that end in the word skill or skills'. This is a simple method that realistically won't cover all possible occurrences of skills, but this is a start."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "skill name: communication skills\n",
      "context: Excellent client presentation and communication skills as well as strong customer service and organizational skills.\n",
      "\n",
      "skill name: organizational skills\n",
      "context: Excellent client presentation and communication skills as well as strong customer service and organizational skills.\n",
      "\n",
      "skill name: communication skills\n",
      "context: We are proud to be an equal opportunity employer College degree preferred, 2 - 5 years experience in print and / or online advertising sales and be able to show consistent sales results in previous positions, Knowledge of the IT industry is preferred, Track record of creativity in sales approaches and solutions, Track record of successfully meeting and exceeding sales goals in media sales relevant to 1105 Medias line of business, Excellent client presentation and communication skills as well as strong customer service and organizational skills, The ideal candidate is energetic, self - motivated, team - oriented, and customer - centric, Understanding of how to research potential customers and use online analytics from a sales perspective, Weekly local travel to meet with clients / prospects is required, Minimal non local travel a few times a year is required\n",
      "\n",
      "skill name: organizational skills\n",
      "context: We are proud to be an equal opportunity employer College degree preferred, 2 - 5 years experience in print and / or online advertising sales and be able to show consistent sales results in previous positions, Knowledge of the IT industry is preferred, Track record of creativity in sales approaches and solutions, Track record of successfully meeting and exceeding sales goals in media sales relevant to 1105 Medias line of business, Excellent client presentation and communication skills as well as strong customer service and organizational skills, The ideal candidate is energetic, self - motivated, team - oriented, and customer - centric, Understanding of how to research potential customers and use online analytics from a sales perspective, Weekly local travel to meet with clients / prospects is required, Minimal non local travel a few times a year is required\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from skills_ml.algorithms.skill_extractors import SkillEndingPatternExtractor\n",
    "from skills_ml.job_postings.common_schema import JobPostingCollectionSample\n",
    "job_posting_generator = JobPostingCollectionSample()\n",
    "# instantiate the skill extractor. This class defaults to only considering lines that\n",
    "# start with a bullet, which doesn't work for this dataset. So we set this flag to False.\n",
    "skill_extractor = SkillEndingPatternExtractor(only_bulleted_lines=False)\n",
    "job_posting = next(iter(job_posting_generator))\n",
    "for candidate_skill in skill_extractor.candidate_skills(job_posting):\n",
    "    print('skill name:', candidate_skill.skill_name)\n",
    "    print('context:', candidate_skill.context)\n",
    "    print('')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The results for one job posting are modest. Two distinct skill names, each occurring two different times in the document.  This is a start.\n",
    "\n",
    "Now let's try another skill extractor: matching with ONET data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "skill name: self\n",
      "context: The ideal candidate is energetic, self-motivated, team-oriented, and customer-centric.\n",
      "\n",
      "skill name: self\n",
      "context: We are proud to be an equal opportunity employer College degree preferred, 2-5 years experience in print and/or online advertising sales and be able to show consistent sales results in previous positions, Knowledge of the IT industry is preferred, Track record of creativity in sales approaches and solutions, Track record of successfully meeting and exceeding sales goals in media sales relevant to 1105 Medias line of business, Excellent client  presentation and communication skills as well as strong customer service and organizational skills, The ideal candidate is energetic, self-motivated, team-oriented, and customer-centric, Understanding of how to research potential customers and use online analytics from a sales perspective, Weekly local travel to meet with clients/prospects is required, Minimal non local travel a few times a year is required\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from skills_ml.algorithms.skill_extractors import ExactMatchSkillExtractor\n",
    "skill_extractor = ExactMatchSkillExtractor(onet.competency_framework)\n",
    "for candidate_skill in skill_extractor.candidate_skills(job_posting):\n",
    "    print('skill name:', candidate_skill.skill_name)\n",
    "    print('context:', candidate_skill.context)\n",
    "    print('')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Yikes. What is this? As it turns out, ['Self' is a real programming language](https://en.wikipedia.org/wiki/Self_(programming_language)). But it's not applicable here. Simply searching for skill names has its limitations.\n",
    "\n",
    "To help with this, there is also the SocScopedExactMatchSkillExtractor. This does exact matching, but only for the occupation that the document is tagged with. This, of course, is only applicable if the document has one. And it needs a full CompetencyOntology to work."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.algorithms.skill_extractors import SocScopedExactMatchSkillExtractor\n",
    "skill_extractor = SocScopedExactMatchSkillExtractor(onet)\n",
    "for candidate_skill in skill_extractor.candidate_skills(job_posting):\n",
    "    print('skill name:', candidate_skill.skill_name)\n",
    "    print('context:', candidate_skill.context)\n",
    "    print('')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "No results. This is expected: For an occupation that is not related to computer programming, the language 'Self' is likely irrelevant.\n",
    "\n",
    "Here's a list of all the other skill extractors available.\n",
    "\n",
    "`FuzzyMatchSkillExtractor` - Similar to the ExactMatchSkillExtractor, but using a configurable edit distance to find skill names that are very close to the targets.\n",
    "\n",
    "`AbilityEndingPatternExtractor` - Similar to the SkillEndingPatternExtractor, but finding noun phrases that end in 'ability' or 'abilities'.\n",
    "\n",
    "`SectionExtractSkillExtractor` - Attempts to divide the text into sections with headers, which is a common pattern found in job postings. Return each individual sentence found in sections with certain headers (Skills, Competencies, Qualifications)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluating Skill Extractors\n",
    "\n",
    "We want to be able to evaluate skill extractors. We may or may not have labeled skills but do want to be able to generate descriptive metrics either way."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "metric: total_candidate_skills value: 153\n",
      "metric: total_vocabulary_size value: 40\n",
      "metric: onet_ksat_competency_recall value: 0.001248868213181804\n"
     ]
    }
   ],
   "source": [
    "from skills_ml.evaluation.skill_extraction_metrics import TotalOccurrences, TotalVocabularySize, OntologyCompetencyRecall\n",
    "\n",
    "metrics = [\n",
    "    TotalOccurrences(),\n",
    "    TotalVocabularySize(),\n",
    "    OntologyCompetencyRecall(onet)\n",
    "]\n",
    "exact_match_skill_extractor = ExactMatchSkillExtractor(onet.competency_framework) \n",
    "for metric in metrics:\n",
    "    candidate_skills = []\n",
    "    for job_posting in job_posting_generator:\n",
    "        candidate_skills += list(exact_match_skill_extractor.candidate_skills(job_posting))\n",
    "    \n",
    "    print('metric:', metric.name, 'value:', metric.eval(candidate_skills, 50)) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Embedding\n",
    "\n",
    "- Labor market data tends to be large in scale, but represented as raw text. Consequently, an important early step for most tasks is to transform texts into a mathematical form that can be used in the downstream tasks.\n",
    "\n",
    "- In the context of skills and jobs, an embedding model trained on large amount of job posting data is able to map a skill or a job title into a high dimensional space as well as preserving the contextual and semantic relationship. Ideally, a good embedding model will cluster similar skills and jobs."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Embedding Models\n",
    "\n",
    "Many word embedding techniques have been developed since the most impactful embedding algorithm word2vec was published in 2013. Currently, Skills-ML includes word2vec, doc2vec and fastext and may include more in the future.\n",
    "\n",
    "- `Word2VecModel` is able to look up a word vector and infer a sentence/paragraph vector by averaging each word in a sentence/paragraph. It supports online learning. For out-of-vocabulary word handling of sentence/paragraph inference, a random vector will be assigned with the same dimension.\n",
    "- `Doc2VecModel` is able to look up a word vector and infer a sentence/paragraph vector by gradient descending on the fly, so it is non-deterministic. It does not support online learning. \n",
    "- `FastTextModel` is able to look up a word vector and infer a sentence/paragraph vector by averaging each word in a sentence/paragraph. It supports online learning. For out-of-vocabulary word handling of sentence/paragraph inference, it sums all vectors of the unseen word’s char-ngrams. If none of the char-ngrams of the unseen word is present, a random vector will be assigned with the same dimension."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.algorithms.embedding.models import Word2VecModel, FastTextModel\n",
    "\n",
    "cbow = Word2VecModel(size=200, sg=0, window=7, iter=3, batch_words=1000)\n",
    "skip_gram = Word2VecModel(size=200, sg=1, window=7, iter=3, batch_words=1000)\n",
    "fasttext = FastTextModel(size=200, window=7, iter=3, batch_words=1000) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Corpora\n",
    "\n",
    "Next, we need some text corpus to train embedding modelss. `Skills-ML` provides pre-defined classes to convert common schema job listings into a corpus in documnet level suitable for use by machine learning algorithms or specific tasks.\n",
    "\n",
    "- `Word2VecGensimCorpusCreator`\n",
    "- `Doc2VecGensimCorpusCreator`\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.job_postings.corpora import Word2VecGensimCorpusCreator, Doc2VecGensimCorpusCreator\n",
    "sampler = JobSampler(job_posting_generator=jobpostings_filtered, k=5000, key=lambda x: x['onet_soc_code'][:2], weights=weights)\n",
    "w2v_corpus_generator = Word2VecGensimCorpusCreator(sampler)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing\n",
    "\n",
    "Or we can build our own corpus generator by using some preprocessing tools\n",
    "\n",
    "Function Compostition\n",
    "- `ProcessingPipeline` will compose processing functions together to become a callable object that takes in the input from the very first processing function and returns the output of the last processing function.\n",
    "- `IterablePipeline` will compose processing functions together to be passed to different stages(training/          prediction)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.algorithms.preprocessing import IterablePipeline\n",
    "from skills_ml.algorithms import nlp\n",
    "from functools import partial\n",
    "\n",
    "document_schema_fields = ['description','experienceRequirements', 'qualifications', 'skills']\n",
    "\n",
    "pipeline = IterablePipeline(\n",
    "    partial(nlp.fields_join, document_schema_fields=document_schema_fields),\n",
    "    nlp.clean_html,\n",
    "    nlp.clean_str,\n",
    "    nlp.word_tokenize,\n",
    ")\n",
    "\n",
    "corpus_generator = pipeline(sampler)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train Embedding\n",
    "\n",
    "The `EmbeddingTrainer` provides online batch learning for Word2VecModel and FastTextModel. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.algorithms.embedding.train import EmbeddingTrainer\n",
    "\n",
    "\n",
    "trainer = EmbeddingTrainer(cbow, skip_gram, fasttext, batch_size=100)\n",
    "trainer.train(corpus_generator)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Storage\n",
    "`Skills-ML` has couple useful storage classes that could benefit both local or cloud. \n",
    "- `S3Store`: S3 storage engine\n",
    "- `FSStore`: File system storage engine\n",
    "- `ModelStorage`: Serialization model storage. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.storage import FSStore, S3Store, ModelStorage\n",
    "\n",
    "fs = FSStore(path=\"tmp/model_cache/embedding/examples\")\n",
    "trainer.save_model(storage=fs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "word2vec_7bdfa911ccc14b971f92b35d529c1dc6.model\n",
      "FSStore(path=tmp/model_cache/embedding/examples)\n"
     ]
    }
   ],
   "source": [
    "print(cbow.model_name)\n",
    "print(cbow.storage)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('developer', 0.7927991151809692) ('developer', 0.6523622870445251)\n",
      "('analyst', 0.7418307662010193) ('tester', 0.624006986618042)\n",
      "('sr', 0.6851106882095337) ('rhev', 0.6191227436065674)\n",
      "('designer', 0.6370913982391357) ('devops', 0.6076809167861938)\n",
      "('devops', 0.6367722749710083) ('modis', 0.5969630479812622)\n",
      "('intern', 0.6265694499015808) ('designer', 0.5858316421508789)\n",
      "('inc', 0.6245660781860352) ('writer', 0.568120002746582)\n",
      "('architect', 0.619118332862854) ('chantilly', 0.5666274428367615)\n",
      "('scientist', 0.6182761192321777) ('island', 0.5646683573722839)\n",
      "('tester', 0.5959925651550293) ('herndon', 0.5613434910774231)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/gensim/matutils.py:737: FutureWarning: Conversion of the second argument of issubdtype from `int` to `np.signedinteger` is deprecated. In future, it will be treated as `np.int64 == np.dtype(int).type`.\n",
      "  if np.issubdtype(vec.dtype, np.int):\n"
     ]
    }
   ],
   "source": [
    "for c, s in zip(cbow.wv.most_similar(['engineer']), skip_gram.wv.most_similar(['engineer'])):\n",
    "    print(c, s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('perl', 0.9427046775817871) ('perl', 0.9121636152267456)\n",
      "('ruby', 0.908499002456665) ('ruby', 0.8980081081390381)\n",
      "('scripting', 0.9076182842254639) ('bash', 0.8887742757797241)\n",
      "('languages', 0.9037582874298096) ('mysql', 0.8292362689971924)\n",
      "('programming', 0.8735607862472534) ('centos', 0.821326494216919)\n",
      "('linux', 0.8646563291549683) ('jsp', 0.8192467093467712)\n",
      "('java', 0.8616237640380859) ('ee', 0.8136248588562012)\n",
      "('javascript', 0.8528693914413452) ('languages', 0.8108319044113159)\n",
      "('css', 0.8422471284866333) ('xml', 0.8019323348999023)\n",
      "('xml', 0.8400859832763672) ('scripting', 0.8000487089157104)\n"
     ]
    }
   ],
   "source": [
    "for c, s in zip(cbow.wv.most_similar(['python']), skip_gram.wv.most_similar(['python'])):\n",
    "    print(c, s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('confined', 0.7270200252532959) ('machining', 0.7596360445022583)\n",
      "('machining', 0.7186369895935059) ('press', 0.7382968664169312)\n",
      "('grinders', 0.6886007785797119) ('igniters', 0.7340461015701294)\n",
      "('graphs', 0.6842017769813538) ('turbines', 0.7310390472412109)\n",
      "('feed', 0.6837735772132874) ('harness', 0.7257314920425415)\n",
      "('lathes', 0.6761114597320557) ('lathes', 0.7163777351379395)\n",
      "('controllers', 0.6728384494781494) ('ntcss', 0.7149447202682495)\n",
      "('wood', 0.6689127683639526) ('comms', 0.7094639539718628)\n",
      "('toxic', 0.6676516532897949) ('xendesktop', 0.7087377905845642)\n",
      "('wire', 0.6662098169326782) ('metal', 0.707713782787323)\n"
     ]
    }
   ],
   "source": [
    "for c, s in zip(cbow.wv.most_similar(['cnc']), skip_gram.wv.most_similar(['cnc'])):\n",
    "    print(c, s)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Skills-ML also provides a function to visualize the embedding in tensorboard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Run `tensorboard --logdir=/home/ubuntu/skills-ml/word2vec_7bdfa911ccc14b971f92b35d529c1dc6 --host 127.0.0.1` to run visualize result on tensorboard\n"
     ]
    }
   ],
   "source": [
    "from skills_ml.algorithms.embedding.models import visualize_in_tensorboard\n",
    "visualize_in_tensorboard(cbow)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation\n",
    "\n",
    "Although there is an emerging trend towards generating embeddings for structured and unstructured data, there is not yet any systematic suite for measuring the quality of embeddings.\n",
    "We generally follow one of the few works in embedding evaluation [Concept2vec: Metrics for Evaluating Quality of Embeddings for Ontological Concepts] to create metrics for evaluating embedding against the gold standard ontology dataset. The gold standard ontology is curated by domain experts like O*NET, so a good embedding should replicate the structure of the entities in the gold standard taxonomy. In other words, it is useful to see how an embedding reflects the clustering structure.\n",
    "\n",
    "One trivial clustering is Major Groups of occupations. A good embedding should cluster the occupations which belong to the same major groups.\n",
    "\n",
    "- `CategorizationMetric`: The cosine similarity between the embedding of the concept and the mean vector of embeddings of all the entities within that concept cluster. This metric aligns a clustering of entities into different categories, reflecting how well the embedding of a concept cluster performs as the background concept of the entities typed by it. \n",
    "\n",
    "- `IntraClusterCohesion`: The sum of squared error of the embedding of the centroid of the concept cluster and the embedding of each entities within that cluster. It measures how near the data points in a cluster are to the cluster centroid.\n",
    "\n",
    "- `MajorGroupRecall`: For a major group, calculate the cosine similarity against all the occupations and find the top n closest occupations. The recall is defined as the number of true positives from top n closest occupations divided by the total number of occupation within the major group.\n",
    "\n",
    "- `MajorGroupPrecision`: Similarly to MajorGroupRecall which is called Coherence Score in the paper, start by finding the top n closest occupations. The precision is defined as the number of true positives from top n closest occupations divided by n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.ontologies.onet import Onet\n",
    "\n",
    "major_group_occupation_des_clustering = onet.major_group_occupation_description_clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.evaluation.embedding_metrics import metrics_for_embedding, CategorizationMetric, IntraClusterCohesion, RecallTopN, PrecisionTopN\n",
    "from skills_ml.algorithms.preprocessing import ProcessingPipeline\n",
    "\n",
    "def vectorization(embedding):\n",
    "    p = ProcessingPipeline(\n",
    "        nlp.normalize,\n",
    "        nlp.clean_str,\n",
    "        nlp.word_tokenize,\n",
    "        partial(nlp.vectorize, embedding_model=embedding)\n",
    "    )\n",
    "    return p\n",
    "\n",
    "categorization_metric = CategorizationMetric(major_group_occupation_des_clustering)\n",
    "intra_cohesion = IntraClusterCohesion(major_group_occupation_des_clustering)\n",
    "recall_top = RecallTopN(major_group_occupation_des_clustering, topn=10)\n",
    "precision_top = PrecisionTopN(major_group_occupation_des_clustering, topn=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Building and Grounds Cleaning and Maintenance': 0.34224581131861653,\n",
       " 'Food Preparation and Serving Related': 0.2654245949239141,\n",
       " 'Construction and Extraction': 0.47868056098318057,\n",
       " 'Healthcare Practitioners and Technical': 0.3940974596927912,\n",
       " 'Transportation and Material Moving': 0.3232632841344041,\n",
       " 'Computer and Mathematical': 0.2728775771822607,\n",
       " 'Life, Physical, and Social Science': 0.5065041654514952,\n",
       " 'Farming, Fishing, and Forestry': 0.46221691564956713,\n",
       " 'Installation, Maintenance, and Repair': 0.18825461785204867,\n",
       " 'Business and Financial Operations': 0.4141572197688639,\n",
       " 'Education, Training, and Library': 0.4499782804964112,\n",
       " 'Military Specific': 0.8406518707733225,\n",
       " 'Production': 0.767452990807525,\n",
       " 'Office and Administrative Support': 0.6310052316123922,\n",
       " 'Healthcare Support': 0.9414934205345233,\n",
       " 'Sales and Related': 0.318627279388384,\n",
       " 'Protective Service': 0.7164475858481743,\n",
       " 'Management': 0.6463669464049742,\n",
       " 'Personal Care and Service': 0.5464152492907202,\n",
       " 'Architecture and Engineering': 0.34074216780013,\n",
       " 'Community and Social Service': 0.39121662365108567,\n",
       " 'Arts, Design, Entertainment, Sports, and Media': 0.4462032355049601,\n",
       " 'Legal': 0.5343651523623081}"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "categorization_metric.eval(vectorization(fasttext))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "import statistics\n",
    "import operator\n",
    "from collections import defaultdict\n",
    "\n",
    "# We define some helper functions to evaluate multiple embeddings\n",
    "def algorithm_name(emb):\n",
    "    if emb.model_type == 'word2vec' or emb.model_type == 'fasttext':\n",
    "        if getattr(emb, 'sg', None) == 1:\n",
    "            return 'Skip-Gram'\n",
    "        else:\n",
    "            return 'Continuous Bag of Words'\n",
    "        \n",
    "    elif emb.model_type == 'doc2vec':\n",
    "        if getattr(emb, 'dm', None) == 1:\n",
    "            return 'Distributed Memory'\n",
    "        else:\n",
    "            return 'Distributed Bag of Words'\n",
    "\n",
    "\n",
    "def evaluate_multiple_embeddings(embeddings, vectorization, metric):\n",
    "    result = defaultdict(dict)\n",
    "    for emb in embeddings:\n",
    "        c = metric.eval(vectorization(emb))\n",
    "        name = emb.model_name.split('.')[0]\n",
    "        result[name]['mean'] = statistics.mean(list(c.values()))\n",
    "        result[name]['variance'] = statistics.variance(list(c.values()))\n",
    "        result[name]['std'] = statistics.stdev(list(c.values()))\n",
    "        result[name]['max'] = max(c.items(), key=operator.itemgetter(1))[1]\n",
    "        result[name]['max_cluster'] = max(c.items(), key=operator.itemgetter(1))[0]\n",
    "        result[name]['min'] = min(c.items(), key=operator.itemgetter(1))[1]\n",
    "        result[name]['min_cluster'] = min(c.items(), key=operator.itemgetter(1))[0]\n",
    "        result[name]['type'] = emb.model_type\n",
    "        result[name]['algorithm'] = algorithm_name(emb)\n",
    "        result[name]['window'] = emb.window\n",
    "    \n",
    "    return pd.DataFrame(result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>word2vec_7bdfa911ccc14b971f92b35d529c1dc6</th>\n",
       "      <th>word2vec_fee333d52ec0101d0e95848a8f376d37</th>\n",
       "      <th>fasttext_6885e864e01553bde10e120d40491de1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>algorithm</th>\n",
       "      <td>Continuous Bag of Words</td>\n",
       "      <td>Skip-Gram</td>\n",
       "      <td>Continuous Bag of Words</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.951364</td>\n",
       "      <td>0.617921</td>\n",
       "      <td>0.941493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max_cluster</th>\n",
       "      <td>Military Specific</td>\n",
       "      <td>Production</td>\n",
       "      <td>Healthcare Support</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.49476</td>\n",
       "      <td>0.349691</td>\n",
       "      <td>0.487769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.149769</td>\n",
       "      <td>0.169084</td>\n",
       "      <td>0.188255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min_cluster</th>\n",
       "      <td>Installation, Maintenance, and Repair</td>\n",
       "      <td>Building and Grounds Cleaning and Maintenance</td>\n",
       "      <td>Installation, Maintenance, and Repair</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.226956</td>\n",
       "      <td>0.119819</td>\n",
       "      <td>0.193501</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type</th>\n",
       "      <td>word2vec</td>\n",
       "      <td>word2vec</td>\n",
       "      <td>fasttext</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>variance</th>\n",
       "      <td>0.0515092</td>\n",
       "      <td>0.0143566</td>\n",
       "      <td>0.0374426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>window</th>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            word2vec_7bdfa911ccc14b971f92b35d529c1dc6  \\\n",
       "algorithm                     Continuous Bag of Words   \n",
       "max                                          0.951364   \n",
       "max_cluster                         Military Specific   \n",
       "mean                                          0.49476   \n",
       "min                                          0.149769   \n",
       "min_cluster     Installation, Maintenance, and Repair   \n",
       "std                                          0.226956   \n",
       "type                                         word2vec   \n",
       "variance                                    0.0515092   \n",
       "window                                              7   \n",
       "\n",
       "                 word2vec_fee333d52ec0101d0e95848a8f376d37  \\\n",
       "algorithm                                        Skip-Gram   \n",
       "max                                               0.617921   \n",
       "max_cluster                                     Production   \n",
       "mean                                              0.349691   \n",
       "min                                               0.169084   \n",
       "min_cluster  Building and Grounds Cleaning and Maintenance   \n",
       "std                                               0.119819   \n",
       "type                                              word2vec   \n",
       "variance                                         0.0143566   \n",
       "window                                                   7   \n",
       "\n",
       "            fasttext_6885e864e01553bde10e120d40491de1  \n",
       "algorithm                     Continuous Bag of Words  \n",
       "max                                          0.941493  \n",
       "max_cluster                        Healthcare Support  \n",
       "mean                                         0.487769  \n",
       "min                                          0.188255  \n",
       "min_cluster     Installation, Maintenance, and Repair  \n",
       "std                                          0.193501  \n",
       "type                                         fasttext  \n",
       "variance                                    0.0374426  \n",
       "window                                              7  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate_multiple_embeddings([cbow, skip_gram, fasttext], vectorization, categorization_metric)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Occupation Classification\n",
    "\n",
    "A common issue with job posting data is incomplete, incorrect, and inconsistent occupation classification. The majority of job postings in the US are using the O*NET SOC classification system, but many are either missing or poorly classified. This can be improved by using machine learning.\n",
    "\n",
    "## SOC Codes\n",
    "Most of the job posting data collected are aligned with the O*NET SOC system. The occupations in the SOC are classified at four levels of aggregation: major group, minor group, broad occupation, and detailed occupation. Each lower level of detail identifies a more specific group of occupations. Each item in the SOC is designated by a six-digit code. The first two digits represent the major group, the third digit represents the minor group, the fourth and fifth digits represent the broad occupation, and the sixth digit represents the detailed occupation.\n",
    "- Major group codes end with 0000 (e.g., 29-0000 Healthcare Practitioners and Technical\n",
    "Occupations —the exceptions are minor groups 15-1200 Computer Occupations, 31- 1100 Home Health and Personal Care Aides; and Nursing Assistants, Orderlies, and Psychiatric Aides, and 51-5100 Printing Workers, which end with 00). \n",
    "- Minor groups generally end with 000 (e.g., 29-1000 Health Diagnosing or Treating Practitioners).\n",
    "- Broad occupations end with 0 (e.g., 29-1020 Dentists).\n",
    "- Detailed occupations end with a number other than 0 (e.g., 29-1022 Oral and Maxillofacial Surgeons).\n",
    "\n",
    "\n",
    "## Target Variable\n",
    "\n",
    "- `FullSOC`\n",
    "- `SOCMajorGroup`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.algorithms.occupation_classifiers import FullSOC, SOCMajorGroup\n",
    "\n",
    "full_soc = FullSOC(onet_cache=onet)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Design Matrix\n",
    "\n",
    "The classification task consists of inferring a SOC code from a job posting and is accomplished through several stages: preprocessing, filtering, training and testing. `DesignMatrix` helps users accomplish this task.\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "from itertools import islice\n",
    "from skills_ml.utils import itershuffle\n",
    "from skills_ml.algorithms.occupation_classifiers import DesignMatrix\n",
    "\n",
    "            \n",
    "sample = JobSampler(job_posting_generator=jobpostings_filtered, k=5000, key=lambda x: x['onet_soc_code'][:2], weights=weights)\n",
    "    \n",
    "dataset = itershuffle(sample)\n",
    "train = islice(dataset, 0, 4000)\n",
    "test = islice(dataset, 4000)\n",
    "            \n",
    "            \n",
    "pipe_x = IterablePipeline(\n",
    "    partial(nlp.fields_join, document_schema_fields=document_schema_fields),\n",
    "    nlp.clean_str,\n",
    "    nlp.word_tokenize,\n",
    "    partial(nlp.vectorize, embedding_model=fasttext)\n",
    ")\n",
    "\n",
    "pipe_y = IterablePipeline(\n",
    "    full_soc.transformer\n",
    ")        \n",
    "    \n",
    "\n",
    "matrix = DesignMatrix(\n",
    "    train, \n",
    "    full_soc,\n",
    "    pipe_x,\n",
    "    pipe_y,\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## OccupationClassifierTrainer\n",
    "`OccupationClassifierTrainer` trains classifiers with cross validation and picks the best classifier with a grid search based on the metric. It takes in a dictionary for the grid search."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skills_ml.algorithms.occupation_classifiers.train import OccupationClassifierTrainer\n",
    "\n",
    "grid_config = {\n",
    "                 'sklearn.ensemble.ExtraTreesClassifier': {\n",
    "                     'n_estimators': [50, 100],\n",
    "                     'criterion': ['entropy', 'gini'],\n",
    "                     'max_depth': [20],\n",
    "                     'max_features': ['log2'],\n",
    "                     'min_samples_split': [10]\n",
    "                      },\n",
    "                 'sklearn.neural_network.MLPClassifier': {\n",
    "                    'hidden_layer_sizes': [100, 500],\n",
    "                     'activation': ['logistic', 'relu'],\n",
    "                     'solver': ['adam']\n",
    "                     },\n",
    "                 }\n",
    "\n",
    "cls_trainer = OccupationClassifierTrainer(\n",
    "    matrix=matrix,\n",
    "    k_folds=3,\n",
    "    grid_config=grid_config,\n",
    "    storage=FSStore('tmp/model_cache/soc_classifiers/examples'),\n",
    "    n_jobs=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/ensemble/weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.\n",
      "  from numpy.core.umath_tests import inner1d\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/model_selection/_split.py:605: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of members in any class cannot be less than n_splits=3.\n",
      "  % (min_groups, self.n_splits)), Warning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/model_selection/_split.py:605: Warning: The least populated class in y has only 1 members, which is too few. The minimum number of members in any class cannot be less than n_splits=3.\n",
      "  % (min_groups, self.n_splits)), Warning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "cls_trainer.train(save=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<ProxyObjectWithStorage at 0x7f568b8360f8 for GridSearchCV at 0x7f5688979358>,\n",
       " <ProxyObjectWithStorage at 0x7f568b38d888 for GridSearchCV at 0x7f5688c20d68>]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cls_trainer.best_estimators"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation\n",
    "\n",
    "Accuracy, recall, precision and f1 are the metrics taken into consideration. Since it is a multi-class classification problem, an overall performance is evaluated by looking at the micro-average and macro-average for the metrics. A macro-average will compute the metric independently for each class and then take the average, whereas a micro-average will aggregate the contributions of all classes and then computes the average. In other words, a macro-average is treating all classes equally. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/ipykernel_launcher.py:14: DeprecationWarning: generator 'itershuffle' raised StopIteration\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "from skills_ml.algorithms.occupation_classifiers.test import OccupationClassifierTester\n",
    "from skills_ml.evaluation.occ_cls_evaluator import OnetOccupationClassificationEvaluator\n",
    "from skills_ml.algorithms.occupation_classifiers.classifiers import CombinedClassifier\n",
    "from skills_ml.algorithms.embedding.train import Reiterable\n",
    "\n",
    "steps = [\n",
    "    partial(nlp.fields_join, document_schema_fields=document_schema_fields),\n",
    "    nlp.normalize,\n",
    "    nlp.clean_str,\n",
    "    nlp.word_tokenize,\n",
    "]\n",
    "\n",
    "evaluators = []\n",
    "test_data = list(test)\n",
    "for cls in cls_trainer.best_estimators:\n",
    "    tester = OccupationClassifierTester(\n",
    "        test_data_generator=test_data, \n",
    "        preprocessing=steps, \n",
    "        classifier=CombinedClassifier(fasttext, cls)\n",
    "    )\n",
    "    evaluators.append(OnetOccupationClassificationEvaluator(tester))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ExtraTreesClassifier(bootstrap=False, class_weight=None, criterion='gini',\n",
      "           max_depth=20, max_features='log2', max_leaf_nodes=None,\n",
      "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "           min_samples_leaf=1, min_samples_split=10,\n",
      "           min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=4,\n",
      "           oob_score=False, random_state=None, verbose=0, warm_start=False)\n",
      "accuracy:  0.478\n",
      "precision:  [0.         0.         0.10891089 ... 0.         0.         0.        ]\n",
      "f1:  [0.         0.         0.18032787 ... 0.         0.         0.        ]\n",
      "major group:  0.554\n",
      "macro precision:  0.3553188734850493\n",
      "micro precision:  0.478\n",
      "recall:  [0.         0.         0.52380952 ... 0.         0.         0.        ]\n",
      "macro recall:  0.2807984722668254\n",
      "micro recall:  0.478\n",
      "macro f1:  0.28787327281642394\n",
      "micro f1:  0.478\n",
      "\n",
      "\n",
      "MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,\n",
      "       beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
      "       hidden_layer_sizes=500, learning_rate='constant',\n",
      "       learning_rate_init=0.001, max_iter=200, momentum=0.9,\n",
      "       nesterovs_momentum=True, power_t=0.5, random_state=None,\n",
      "       shuffle=True, solver='adam', tol=0.0001, validation_fraction=0.1,\n",
      "       verbose=False, warm_start=False)\n",
      "accuracy:  0.514\n",
      "precision:  [0.         0.         0.14285714 ... 0.         0.         0.        ]\n",
      "f1:  [0.  0.  0.2 ... 0.  0.  0. ]\n",
      "major group:  0.618\n",
      "macro precision:  0.2966284764221296\n",
      "micro precision:  0.514\n",
      "recall:  [0.         0.         0.33333333 ... 0.         0.         0.        ]\n",
      "macro recall:  0.2887478119504879\n",
      "micro recall:  0.514\n",
      "macro f1:  0.27660324759833993\n",
      "micro f1:  0.514\n",
      "\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.\n",
      "  if diff:\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.\n",
      "  if diff:\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/metrics/classification.py:1135: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/metrics/classification.py:1135: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/metrics/classification.py:1137: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
      "  'recall', 'true', average, warn_for)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/metrics/classification.py:1137: UndefinedMetricWarning: Recall is ill-defined and being set to 0.0 in labels with no true samples.\n",
      "  'recall', 'true', average, warn_for)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.\n",
      "  if diff:\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/preprocessing/label.py:151: DeprecationWarning: The truth value of an empty array is ambiguous. Returning False, but in future this will result in an error. Use `array.size > 0` to check that an array is not empty.\n",
      "  if diff:\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/metrics/classification.py:1135: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/metrics/classification.py:1135: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/metrics/classification.py:1137: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
      "  'recall', 'true', average, warn_for)\n",
      "/home/ubuntu/.pyenv/versions/3.6.5/envs/env3.6.5/lib/python3.6/site-packages/sklearn/metrics/classification.py:1137: UndefinedMetricWarning: Recall is ill-defined and being set to 0.0 in labels with no true samples.\n",
      "  'recall', 'true', average, warn_for)\n"
     ]
    }
   ],
   "source": [
    "for e, c in zip(evaluators, cls_trainer.best_estimators):\n",
    "    print(c.best_estimator_)\n",
    "    print('accuracy: ', e.accuracy)\n",
    "    print('precision: ', e.precision)\n",
    "    print('f1: ', e.f1)\n",
    "    print('major group: ', e.accuracy_major_group)\n",
    "    print('macro precision: ', e.macro_precision)\n",
    "    print('micro precision: ', e.micro_precision)\n",
    "    print('recall: ', e.recall)\n",
    "    print('macro recall: ', e.macro_recall)\n",
    "    print('micro recall: ', e.micro_recall)\n",
    "    print('macro f1: ', e.macro_f1)\n",
    "    print('micro f1: ', e.micro_f1)\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}