Wk3_Submission.ipynb

wk3_submission.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Wk3_Submission.ipynb",
      "provenance": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/patternproject/992663bce97c95dbdf2ec0c76b2ae08f/wk3_submission.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ExrcEKOJqRMS",
        "colab_type": "text"
      },
      "source": [
        "Manning LP \"Using Online Job Postings to Improve Your Data Science Resume\" "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GTXpVFDmqdGi",
        "colab_type": "text"
      },
      "source": [
        "Week 3 - Clustering Job Posting Skill Requirements"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-GSEEXdWOKpl",
        "colab_type": "text"
      },
      "source": [
        "# 3.5 Action Starts Here ... "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "X1LYPAywxnbp",
        "colab_type": "text"
      },
      "source": [
        "## 1.Importing Libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xXHEd5UKxnFe",
        "colab_type": "code",
        "outputId": "b8be10bd-8856-4b03-e1ca-8987abbbb52d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        }
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "\n",
        "# string processing, reg expressions\n",
        "import re, string\n",
        "\n",
        "# for freq vectorizer\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "\n",
        "# for DIM Red\n",
        "from sklearn.decomposition import TruncatedSVD\n",
        "\n",
        "from sklearn.pipeline import make_pipeline\n",
        "from sklearn.preprocessing import Normalizer\n",
        "\n",
        "# for Metrics\n",
        "from sklearn import metrics\n",
        "from sklearn.metrics import silhouette_score, silhouette_samples\n",
        "\n",
        "# for Clustering\n",
        "from sklearn.cluster import KMeans\n",
        "\n",
        "# for Text Processing \n",
        "import nltk\n",
        "from nltk.stem import PorterStemmer\n",
        "from nltk.corpus import stopwords\n",
        "nltk.download('stopwords')\n",
        "nltk.download('punkt')\n",
        "\n",
        "from pylab import *\n",
        "\n",
        "# used in elbow plot\n",
        "import scipy.spatial.distance as scdist\n",
        "\n",
        "from collections import Counter\n",
        "\n",
        "from wordcloud import WordCloud\n"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Package stopwords is already up-to-date!\n",
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Package punkt is already up-to-date!\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GBWZzknYOPLJ",
        "colab_type": "text"
      },
      "source": [
        "## 2.Loading the Dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "asdB9SEnOOs-",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_pkl = pd.read_pickle('step2_df.pk')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KQ3RKcloxzw9",
        "colab_type": "code",
        "outputId": "6ee44980-a10e-44a8-9d21-a7dcfc4e60d2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197
        }
      },
      "source": [
        "df_pkl.head()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>title</th>\n",
              "      <th>body</th>\n",
              "      <th>bullets</th>\n",
              "      <th>cosine_similarity</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Institutional Data and Research Analyst (6948U...</td>\n",
              "      <td>Institutional Data and Research Analyst (6948U...</td>\n",
              "      <td>()</td>\n",
              "      <td>0.143349</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Data Science Health Innovation Fellow Job - BI...</td>\n",
              "      <td>Data Science Health Innovation Fellow Job - BI...</td>\n",
              "      <td>(Demonstrated ability to propose, initiate, an...</td>\n",
              "      <td>0.125523</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Machine Learning Postdoctoral Fellow - San Fra...</td>\n",
              "      <td>Machine Learning Postdoctoral Fellow - San Fra...</td>\n",
              "      <td>(Design and develop distributed machine learni...</td>\n",
              "      <td>0.121162</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Data Analyst (6256U) 1737 - 1737 - Berkeley, C...</td>\n",
              "      <td>Data Analyst (6256U) 1737 - 1737 - Berkeley, C...</td>\n",
              "      <td>()</td>\n",
              "      <td>0.117481</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Senior Data Systems Analyst (0599U) - 1668 - 1...</td>\n",
              "      <td>Senior Data Systems Analyst (0599U) - 1668 - 1...</td>\n",
              "      <td>()</td>\n",
              "      <td>0.113083</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                               title  ... cosine_similarity\n",
              "0  Institutional Data and Research Analyst (6948U...  ...          0.143349\n",
              "1  Data Science Health Innovation Fellow Job - BI...  ...          0.125523\n",
              "2  Machine Learning Postdoctoral Fellow - San Fra...  ...          0.121162\n",
              "3  Data Analyst (6256U) 1737 - 1737 - Berkeley, C...  ...          0.117481\n",
              "4  Senior Data Systems Analyst (0599U) - 1668 - 1...  ...          0.113083\n",
              "\n",
              "[5 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xgZXH3gfx-up",
        "colab_type": "code",
        "outputId": "bdd2dd12-a5cb-4c8f-f25d-70fab8acdea2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        }
      },
      "source": [
        "print(df_pkl.describe)"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<bound method NDFrame.describe of                                                 title  ... cosine_similarity\n",
            "0   Institutional Data and Research Analyst (6948U...  ...          0.143349\n",
            "1   Data Science Health Innovation Fellow Job - BI...  ...          0.125523\n",
            "2   Machine Learning Postdoctoral Fellow - San Fra...  ...          0.121162\n",
            "3   Data Analyst (6256U) 1737 - 1737 - Berkeley, C...  ...          0.117481\n",
            "4   Senior Data Systems Analyst (0599U) - 1668 - 1...  ...          0.113083\n",
            "..                                                ...  ...               ...\n",
            "70              Data Scientist - Pittsburgh, PA 15206  ...          0.057230\n",
            "71  AI Research Scientist - Natural Language Proce...  ...          0.056983\n",
            "72   Senior Data Scientist - Authorship - Oakland, CA  ...          0.056819\n",
            "73              Senior Data Scientist - Palo Alto, CA  ...          0.056781\n",
            "74                  Data Scientist - Denver, CO 80221  ...          0.056697\n",
            "\n",
            "[75 rows x 4 columns]>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IpfKfXv7yFis",
        "colab_type": "code",
        "outputId": "de452be5-2472-4d79-b353-b21fd091c4e9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 105
        }
      },
      "source": [
        "df_pkl.dtypes"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "title                 object\n",
              "body                  object\n",
              "bullets               object\n",
              "cosine_similarity    float64\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_0N5_-yfyMn2",
        "colab_type": "code",
        "outputId": "72ab14ee-6db7-45f9-e516-f17dcfdf8992",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 176
        }
      },
      "source": [
        "df_pkl.info()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 75 entries, 0 to 74\n",
            "Data columns (total 4 columns):\n",
            "title                75 non-null object\n",
            "body                 75 non-null object\n",
            "bullets              75 non-null object\n",
            "cosine_similarity    75 non-null float64\n",
            "dtypes: float64(1), object(3)\n",
            "memory usage: 2.5+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CxCcmszPyXq4",
        "colab_type": "code",
        "outputId": "a75f4614-4560-4db5-bc6c-11f8de6b4f82",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 105
        }
      },
      "source": [
        "df_pkl.infer_objects().dtypes"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "title                 object\n",
              "body                  object\n",
              "bullets               object\n",
              "cosine_similarity    float64\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xdln35hF0d2y",
        "colab_type": "code",
        "outputId": "f985f922-9efb-4d58-e109-69d89ce1c164",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 303
        }
      },
      "source": [
        "df_pkl['bullets'][1]"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "('Demonstrated ability to propose, initiate, and carry out ambitious data-intensive research projects.',\n",
              " 'Entrepreneurial abilities: demonstrated skills in finding unmet needs, translating those into tractable solutions that can be implemented, and working with few specifically assigned resources.',\n",
              " 'Self-motivated and works well both independently and as part of a team.',\n",
              " 'Strong collaboration skills with highly technical researchers and ability to engage across a variety of methodological fields (e.g., computer science, mathematics, and statistics), computational platforms, and ideally, research domains (e.g., health sciences, life sciences, social sciences, and medicine).',\n",
              " 'Excellent and demonstrated ability to regularly, effectively communicate with management teams.',\n",
              " 'Ability to communicate data insights that translate into significant impact in a clear and effective manner to technical and non-technical personnel at various levels in the organization and to external research and education audiences.',\n",
              " 'In depth skills and experience with independently resolving complex computing / data / CI problems using introductory and / or intermediate principles.',\n",
              " 'Ability to curate/clean/organize large and messy datasets; to write code to query and transform both unstructured and structured data.',\n",
              " 'Strong programming skills in scripting languages such as Python, Java/Scala, and SQL and comfort with advanced analytics tools such as R, Spark, and/or Tableau, in addition to in programming languages (e.g. C/C++). Strong working knowledge of Hadoop, extraction/transformation/loading (ETL). Record of prototyping, developing and scaling software.',\n",
              " 'Fluent in using scientific computing environments, e.g. HPC cluster (CPUs and GPUs) or cloud (AWS, Azure, Google, Salesforce, IBM, or VMWare).',\n",
              " 'Highly advanced skills, and extensive experience associated multiple of the following: data modeling; data mining; mathematical modeling; artificial intelligence; machine learning; deep learning models (e.g., 2/3d CNN, LSTM/GRU), architecture (e.g., Resnet, U-net), and frameworks (e.g.,Tensorflow, pytorch, keras); natural language processing; knowledge graphs; reinforcement learning; data representation, and optimization.',\n",
              " 'BS with 7+ years or MS with 6+ years or PhD with 3+ years of applicable experience is expected. Degree(s) should be in a technical discipline such as Computer Science, Engineering, Statistics, Physic, Math or other related fields.',\n",
              " 'Experience working in health and biomedical sectors preferred, but not required.',\n",
              " 'This is a two-year contract position. Contract positions may be extended based on operational demand. Contract positions are eligible to participate in the health and welfare programs offered by UC Berkeley.',\n",
              " 'The salary range designated for this position: $117,800 - $130,000; however, starting salary will be commensurate with experience.')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oZmSDyy0_tSm",
        "colab_type": "text"
      },
      "source": [
        "## 3.Helper Functions"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MSDwxhTw_t0h",
        "colab_type": "code",
        "outputId": "8e16fa58-a431-4d31-c2c1-ef28ae699fc1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "# Python3 code to convert tuple  \n",
        "# into string \n",
        "def convertTuple(tup): \n",
        "    str =  ''.join(tup) \n",
        "    return str\n",
        "  \n",
        "# Driver code \n",
        "tuple = ('g', 'e', 'e', 'k', 's') \n",
        "str = convertTuple(tuple) \n",
        "print(str) "
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "geeks\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AuEfyIM36gQv",
        "colab_type": "text"
      },
      "source": [
        "Function for text processing, used in TF-IDF Pre-processing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6sG4jedW_yUe",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def pre_process(text):\n",
        "    \n",
        "    # lowercase\n",
        "    text=text.lower()\n",
        "    \n",
        "    #remove tags\n",
        "    #text=re.sub(\"<!--?.*?-->\",\"\",text)\n",
        "    \n",
        "    # remove special characters and digits\n",
        "    text=re.sub(\"(\\\\d|\\\\W)+\",\" \",text)\n",
        "\n",
        "    # stemming\n",
        "    ## create an object of class PorterStemmer\n",
        "    porter = PorterStemmer()\n",
        "    text = porter.stem(text)\n",
        "\n",
        "    # removing stop words\n",
        "    #text = ''.join(w for w in text if not w in stopwords.words('english'))\n",
        "    temp = ''.join(w for w in text if not w in stopwords.words('english'))\n",
        "    #print(text)\n",
        "    #print(temp)\n",
        "        \n",
        "    return text"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "89uHGtQe6oDC",
        "colab_type": "text"
      },
      "source": [
        "Helper Function for Color Palette"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4-_ss0Z66ZPt",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def _color_func(*args, **kwargs): \n",
        "  return np.random.choice(['black', 'blue', 'teal', 'purple', 'brown'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3VkvlxCR6wex",
        "colab_type": "text"
      },
      "source": [
        "Workhorse - Generates Word Cloud for a Given Cluster"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6U2IMldW6eZa",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def cluster_to_image(df_cluster, max_words=15): \n",
        "  \n",
        " \n",
        "  # From our previous IDF values, compute the tf-idf scores for documents within this cluster\n",
        "  s_temp = df_cluster['text'].str.cat(sep=',')\n",
        "  s_temp=[s_temp]\n",
        "  tfidf_matrix_all = tfidf_vectorizer.fit_transform(s_temp)\n",
        "\n",
        "  num_samples, num_features = tfidf_matrix_all.shape\n",
        "  print(\"#samples: %d, #features: %d\" % (num_samples, num_features))\n",
        " \n",
        "  #get the top n scores\n",
        "  df = pd.DataFrame(tfidf_matrix_all.T.todense(), index=tfidf_vectorizer.get_feature_names(), columns=[\"tfidf\"])\n",
        "  df_top_n = df.sort_values(by=[\"tfidf\"],ascending=False).head(max_words)\n",
        "  df_top_n = df_top_n.apply(lambda x: np.round(x, decimals=2))\n",
        "  #df_top_n\n",
        "  \n",
        "  # setting index name explicity, required by to_dict()\n",
        "  df_top_n.index.name = 'feat'\n",
        "\n",
        "  # getting a dictionary from df\n",
        "  dict_from_df = df_top_n.T.to_dict('list')\n",
        "\n",
        "  # from list to float\n",
        "  words_to_score = {k:v[0] for k, v in dict_from_df.items()}\n",
        "  #words_to_score\n",
        "\n",
        "  cloud_generator = WordCloud(background_color='white', color_func=_color_func, random_state=1)\n",
        "  \n",
        "  wordcloud_image = cloud_generator.fit_words(words_to_score)\n",
        "  \n",
        "  return wordcloud_image"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mtrYnoKM_-Ha",
        "colab_type": "text"
      },
      "source": [
        "## 4.Some Preprocessing"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gBsIFMkvDY5O",
        "colab_type": "text"
      },
      "source": [
        "convert tuples to strings in bullets column"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zGQoiJYpDbjQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_pkl['bull_strings'] = df_pkl['bullets'].map(convertTuple)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0vLu3StbDYCg",
        "colab_type": "code",
        "outputId": "21781886-59b7-48bb-9957-07818f9d0206",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 194
        }
      },
      "source": [
        "df_pkl.info()"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 75 entries, 0 to 74\n",
            "Data columns (total 5 columns):\n",
            "title                75 non-null object\n",
            "body                 75 non-null object\n",
            "bullets              75 non-null object\n",
            "cosine_similarity    75 non-null float64\n",
            "bull_strings         75 non-null object\n",
            "dtypes: float64(1), object(4)\n",
            "memory usage: 3.1+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CWwb1anN-j--",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_idf = df_pkl \n",
        "df_idf['text'] = df_idf['body'] + df_idf['bull_strings']"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ycVJFYWI-PIf",
        "colab_type": "code",
        "outputId": "91d35ba9-0aa3-43d1-9938-d778396592d2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "source": [
        "df_idf['text'] = df_idf['text'].apply(lambda x:pre_process(x))\n",
        " \n",
        "df_idf['text'][2] "
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'machine learning postdoctoral fellow san francisco bay area ca berkeley lab has an opening for a postdoctoral scholar to work in an emerging new area on novel machine learning and control techniques being applied to computer infrastructure e g networking compute etc this work will lead to the development of automation and optimization techniques for systems to become self aware and intelligent the goal of these networks is to effectively support complex distributed science big data needs across complex multi domain infrastructure such as department of energy doe facilities developments of these techniques will lead to immense impact in the next generation of scientific discoveries requiring massive amounts of data movement from a variety of sources and emerging new trends with exascale computing leveraging techniques from deep learning pattern recognition and even artificial intelligence as a whole this position will explore developing novel techniques deep reinforcement learning working with large distributed data sets and quick processing of the model results the successful candidate will work as part of unique and engaging team developing techniques linking two major research themes namely distributed deep reinforcement learning methods and real world translation to doe applications this work will lead to impactful results advancing state of the art network research by building novel deep reinforcement learning what you will do design and develop distributed machine learning approaches that are applicable to systems network compute facilities environments develop and use software libraries to process large data sets that are publicly and privately available research and implement machine and reinforcement learning models that connect fundamental engineering properties to efficient designs develop new tools for innovative network and compute solutions that leverage machine learning as part of their design apply distributed machine learning techniques to related fields publish research findings in conferences and journals what is required ph d in related field computer science control engineering any engineering discipline electronic and controls mathematics strong analytical background in machine learning algorithms e g experience with deep learning analysis and programming libraries strong mathematical foundation extremely high aptitude and desire for programming and building software infrastructure object oriented programming machine learning libraries prior ability demonstrated e g github profile to contribute to a large software framework demonstrate some history with working with either of the machine learning libraries such as tensorflow pytorch keras scikit learn etc preferred language of experience is python but flexible to be extended to matlab or other machine learning coding experience additional desired qualifications prior work in the field of deep learning research is a plus prior experience in systems automation tools is desirable prior experience of deploying software architectures on hpc and understanding their use is a plus any experience with processing streaming large data sets such as over cloud infrastructures and demonstration using software tools is desirable the posting shall remain open until the position is filled notes this is a full time year postdoctoral appointment with the possibility of renewal based upon satisfactory job performance continuing availability of funds and ongoing operational needs you must have less than years paid postdoctoral experience salary for postdoctoral positions depends on years of experience post degree this position is represented by a union for collective bargaining purposes salary will be predetermined based on postdoctoral step rates this position may be subject to a background check any convictions will be evaluated to determine if they directly relate to the responsibilities and requirements of the position having a conviction history will not automatically disqualify an applicant from being considered for employment work will be primarily performed at lawrence berkeley national lab cyclotron road berkeley ca learn about us berkeley lab lbnl addresses the world s most urgent scientific challenges by advancing sustainable energy protecting human health creating new materials and revealing the origin and fate of the universe founded in berkeley lab s scientific expertise has been recognized with nobel prizes the university of california manages berkeley lab for the u s department of energy s office of science working at berkeley lab has many rewards including a competitive compensation program excellent health and welfare programs a retirement program that is second to none and outstanding development opportunities to view information about the many rewards that are offered at berkeley lab click here equal employment opportunity berkeley lab is an equal opportunity affirmative action employer all qualified applicants will receive consideration for employment without regard to race color religion sex sexual orientation gender identity national origin disability age or protected veteran status berkeley lab is in compliance with the pay transparency nondiscrimination provision under cfr click here to view the poster and supplement equal employment opportunity is the law lawrence berkeley national laboratory encourages applications from women minorities veterans and other underrepresented groups presently considering scientific research careers design and develop distributed machine learning approaches that are applicable to systems network compute facilities environments develop and use software libraries to process large data sets that are publicly and privately available research and implement machine and reinforcement learning models that connect fundamental engineering properties to efficient designs develop new tools for innovative network and compute solutions that leverage machine learning as part of their design apply distributed machine learning techniques to related fields publish research findings in conferences and journals ph d in related field computer science control engineering any engineering discipline electronic and controls mathematics strong analytical background in machine learning algorithms e g experience with deep learning analysis and programming libraries strong mathematical foundation extremely high aptitude and desire for programming and building software infrastructure object oriented programming machine learning libraries prior ability demonstrated e g github profile to contribute to a large software framework demonstrate some history with working with either of the machine learning libraries such as tensorflow pytorch keras scikit learn etc preferred language of experience is python but flexible to be extended to matlab or other machine learning coding experience prior work in the field of deep learning research is a plus prior experience in systems automation tools is desirable prior experience of deploying software architectures on hpc and understanding their use is a plus any experience with processing streaming large data sets such as over cloud infrastructures and demonstration using software tools is desirable this is a full time year postdoctoral appointment with the possibility of renewal based upon satisfactory job performance continuing availability of funds and ongoing operational needs you must have less than years paid postdoctoral experience salary for postdoctoral positions depends on years of experience post degree this position is represented by a union for collective bargaining purposes salary will be predetermined based on postdoctoral step rates this position may be subject to a background check any convictions will be evaluated to determine if they directly relate to the responsibilities and requirements of the position having a conviction history will not automatically disqualify an applicant from being considered for employment work will be primarily performed at lawrence berkeley national lab cyclotron road berkeley ca '"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Xx-ZzMp9_Ktc",
        "colab_type": "text"
      },
      "source": [
        "Drop unnecessary columns"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RizePnIl_RTv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_idf.drop(columns=['bull_strings'],inplace=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SyChwVy7NoWM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 194
        },
        "outputId": "3a4e8376-f43b-48f8-dd72-e37147aee9c0"
      },
      "source": [
        "df_idf.info()"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 75 entries, 0 to 74\n",
            "Data columns (total 5 columns):\n",
            "title                75 non-null object\n",
            "body                 75 non-null object\n",
            "bullets              75 non-null object\n",
            "cosine_similarity    75 non-null float64\n",
            "text                 75 non-null object\n",
            "dtypes: float64(1), object(4)\n",
            "memory usage: 3.1+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iAKx13bA_jsY",
        "colab_type": "text"
      },
      "source": [
        "## 5.Text Normalization \n",
        "\n",
        "---\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TsVQRp4U6ANK",
        "colab_type": "text"
      },
      "source": [
        "#### Word Frequencies with TfidfVectorizer\n",
        "\n",
        "Transform job posting text and our resume into TFIDF vectors using sklearn’s TFIDF vectorizer class.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yJhFtonfcjfR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "stemmer = PorterStemmer()\n",
        "\n",
        "def stem_words(words_list, stemmer):\n",
        "    return [stemmer.stem(word) for word in words_list]\n",
        "\n",
        "def tokenize(text):\n",
        "    tokens = nltk.word_tokenize(text)\n",
        "    stems = stem_words(tokens, stemmer)\n",
        "    return stems"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cNt9B8fAEO9z",
        "colab_type": "text"
      },
      "source": [
        "   - Strips out “stop words” (stop_words='english')\n",
        "   - Filters out terms that occur in more than half of the docs (max_df=0.5) >> Not Used\n",
        "   - Filters out terms that occur in only one document (min_df=2). >> Not Used, Was giving error\n",
        "   - Selects the 10,000 most frequently occuring words in the corpus. (max_features=10000)\n",
        "   - Normalizes the vector (L2 norm of 1.0) to normalize the effect of \n",
        "     document length on the tf-idf values. (use_idf=True)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RI1G2QKq_un_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# create the transform\n",
        "\n",
        "tfidf_vectorizer = TfidfVectorizer(stop_words='english', max_features=10000, use_idf=True)  \n",
        "\n",
        "# tokenize and build vocab\n",
        "tfidf_matrix = tfidf_vectorizer.fit_transform(df_idf['text'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gC6Z680GOLQ5",
        "colab_type": "text"
      },
      "source": [
        "Compare the tfidf dimensions with orginial dataframe (from 75 x 1) to (75 x 4415)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OvfRu2Dc5AyX",
        "colab_type": "code",
        "outputId": "2e020066-0210-4b57-cc42-4d18a74c4928",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "num_samples, num_features = tfidf_matrix.shape\n",
        "\n",
        "print(\"#samples: %d, #features: %d\" % (num_samples, num_features))\n"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "#samples: 75, #features: 4415\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J8w1huNwLxT_",
        "colab_type": "code",
        "outputId": "f5f78cbb-e228-4aa7-f299-ab0dc4150ea1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        }
      },
      "source": [
        "print(repr(tfidf_matrix))"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<75x4415 sparse matrix of type '<class 'numpy.float64'>'\n",
            "\twith 18625 stored elements in Compressed Sparse Row format>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2ZemOBMD_2Iu",
        "colab_type": "code",
        "outputId": "3b450969-bd6f-458d-cc05-092e916197f3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "source": [
        "# summarize\n",
        "print(tfidf_vectorizer.vocabulary_)"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'institutional': 2003, 'data': 954, 'research': 3424, 'analyst': 177, 'berkeley': 405, 'ca': 483, 'university': 4210, 'california': 488, 'world': 4387, 'iconic': 1878, 'teaching': 4009, 'institutions': 2004, 'fueled': 1643, 'perpetual': 2869, 'renaissance': 3383, 'generating': 1685, 'unparalleled': 4219, 'intellectual': 2024, 'economic': 1238, 'social': 3742, 'value': 4267, 'united': 4206, 'states': 3845, 'culture': 918, 'openness': 2707, 'freedom': 1630, 'acceptance': 17, 'academic': 10, 'artistic': 270, 'political': 2969, 'cultural': 917, 'make': 2353, 'special': 3775, 'place': 2933, 'students': 3894, 'faculty': 1510, 'staff': 3823, 'committed': 712, 'hiring': 1823, 'developing': 1086, 'want': 4334, 'work': 4376, 'high': 1814, 'performing': 2862, 'supports': 3952, 'outstanding': 2760, 'deciding': 1000, 'apply': 229, 'position': 2982, 'candidates': 501, 'strongly': 3888, 'encouraged': 1312, 'consider': 791, 'alignment': 146, 'workplace': 4385, 'potential': 3000, 'success': 3918, 'http': 1859, 'jobs': 2124, 'edu': 1250, 'html': 1858, 'application': 222, 'review': 3480, 'date': 983, 'job': 2121, 'september': 3645, 'departmental': 1039, 'overview': 2769, 'graduate': 1731, 'division': 1170, 'oversees': 2767, 'stage': 3825, 'student': 3893, 'life': 2270, 'applications': 224, 'admissions': 84, 'fellowships': 1536, 'grants': 1737, 'tracking': 4114, 'progress': 3137, 'degree': 1016, 'milestones': 2482, 'employment': 1300, 'preparation': 3038, 'mentoring': 2441, 'activities': 56, 'professional': 3117, 'development': 1087, 'mission': 2502, 'sustain': 3959, 'renowned': 3386, 'excellence': 1433, 'master': 2395, 'doctoral': 1179, 'programs': 3135, 'assist': 289, 'financial': 1562, 'personal': 2871, 'matters': 2407, 'studies': 3897, 'implements': 1915, 'policies': 2966, 'developed': 1084, 'council': 868, 'senate': 3635, 'ensure': 1344, 'quality': 3227, 'education': 1252, 'distribute': 1163, 'resources': 3440, 'equitably': 1384, 'disciplines': 1134, 'schools': 3576, 'colleges': 683, 'graduates': 1732, 'recipients': 3307, 'annually': 195, 'including': 1932, 'populations': 2976, 'underrepresented': 4195, 'higher': 1815, 'does': 1187, 'numerous': 2658, 'cutting': 936, 'edge': 1246, 'attract': 314, 'diverse': 1167, 'backgrounds': 357, 'passions': 2826, 'support': 3950, 'makes': 2356, 'second': 3605, 'breadth': 453, 'depth': 1052, 'reach': 3272, 'scholarship': 3574, 'public': 3187, 'service': 3657, 'impact': 1906, 'globe': 1706, 'responsibilities': 3447, 'researches': 3427, 'gathers': 1672, 'analyzes': 185, 'visualizes': 4317, 'complex': 748, 'reports': 3395, 'projects': 3140, 'using': 4251, 'tableau': 3980, 'statistical': 3850, 'packages': 2782, 'presents': 3057, 'findings': 1565, 'recommendations': 3315, 'clear': 624, 'concise': 771, 'manner': 2372, 'access': 19, 'multiple': 2561, 'sources': 3768, 'oracle': 2729, 'bi': 410, 'sql': 3809, 'contributes': 830, 'design': 1060, 'deployment': 1048, 'documentation': 1182, 'systems': 3976, 'structures': 3891, 'tools': 4103, 'visualization': 4312, 'information': 1964, 'collection': 678, 'retrieval': 3471, 'reporting': 3394, 'summarize': 3928, 'creative': 898, 'non': 2625, 'technical': 4017, 'presentation': 3052, 'website': 4348, 'uses': 4250, 'produce': 3105, 'routine': 3519, 'answer': 199, 'ad': 62, 'hoc': 1830, 'queries': 3236, 'campus': 495, 'oversee': 2765, 'study': 3900, 'required': 3414, 'qualifications': 3222, 'requires': 3421, 'skill': 3720, 'software': 3745, 'analysis': 167, 'skills': 3722, 'experience': 1456, 'visual': 4310, 'query': 3237, 'ability': 2, 'systematically': 3975, 'transform': 4131, 'connect': 786, 'disparate': 1151, 'sets': 3664, 'general': 1680, 'knowledge': 2177, 'science': 3578, 'methods': 2458, 'principles': 3069, 'procedures': 3089, 'involved': 2081, 'handling': 1781, 'sensitive': 3642, 'learn': 2220, 'example': 1427, 'family': 1517, 'educational': 1253, 'rights': 3492, 'privacy': 3077, 'act': 48, 'ferpa': 1538, 'understanding': 4198, 'rules': 3525, 'regulations': 3359, 'active': 54, 'listening': 2296, 'critical': 907, 'thinking': 4070, 'good': 1711, 'interpersonal': 2052, 'multi': 2554, 'task': 3998, 'demonstrated': 1033, 'tell': 4034, 'compelling': 729, 'story': 3871, 'visually': 4319, 'writing': 4393, 'training': 4122, 'bachelor': 353, 'computer': 763, 'statistics': 3851, 'new': 2602, 'media': 2421, 'hci': 1795, 'interaction': 2031, 'equivalent': 1386, 'preferred': 3028, 'strong': 3887, 'portfolio': 2979, 'years': 4405, 'combined': 694, 'formal': 1607, 'prior': 3072, 'desirable': 1068, 'salary': 3537, 'benefits': 401, 'exempt': 1443, 'monthly': 2543, 'paid': 2785, 'annual': 194, 'commensurate': 704, 'range': 3258, 'comprehensive': 758, 'package': 2781, 'offered': 2681, 'visit': 4306, 'ucnet': 4181, 'universityofcalifornia': 4211, 'compensation': 731, 'index': 1946, 'health': 1798, 'innovation': 1981, 'fellow': 1533, 'bids': 413, 'immediate': 1902, 'uc': 4178, 'ucsf': 4183, 'johnson': 2125, 'janssen': 2110, 'came': 493, 'establish': 1395, 'fellowship': 1535, 'program': 3130, 'bring': 461, 'experienced': 1458, 'scientists': 3584, 'industry': 1955, 'researchers': 3426, 'facilitate': 1503, 'collaboration': 667, 'engineers': 1329, 'clinicians': 639, 'entrepreneurs': 1359, 'tackle': 3982, 'problems': 3085, 'patient': 2833, 'care': 513, 'accelerate': 13, 'healthcare': 1799, 'driven': 1208, 'approaches': 238, 'offers': 2682, 'year': 4404, 'entrepreneurial': 1358, 'opportunity': 2718, 'seasoned': 3603, 'professionals': 3119, 'engineering': 1327, 'computational': 760, 'relevant': 3374, 'domain': 1190, 'datasets': 975, 'advance': 89, 'improve': 1921, 'providing': 3177, 'meaningful': 2412, 'clinical': 638, 'insights': 1992, 'recruit': 3327, 'serial': 3650, 'cohorts': 663, 'academia': 9, 'seeking': 3615, 'wide': 4364, 'specializations': 3779, 'burgeoning': 478, 'field': 1541, 'artificial': 269, 'intelligence': 2026, 'cloud': 647, 'computing': 764, 'optimizing': 2724, 'methodologies': 2456, 'working': 4383, 'various': 4273, 'large': 2198, 'fellows': 1534, 'worked': 4378, 'related': 3366, 'electronic': 1276, 'medical': 2422, 'records': 3326, 'notes': 2645, 'images': 1899, 'content': 817, 'microscopy': 2473, 'omics': 2693, 'comparable': 727, 'levels': 2256, 'scale': 3561, 'complexity': 749, 'receive': 3299, 'guidance': 1763, 'experts': 1473, 'mentor': 2439, 'collaborating': 666, 'organizations': 2738, 'primary': 3067, 'responsibility': 3448, 'successfully': 3920, 'identify': 1886, 'conduct': 775, 'interdisciplinary': 2035, 'intensive': 2029, 'use': 4245, 'technology': 4030, 'tool': 4100, 'accomplish': 25, 'aims': 128, 'considered': 794, 'applicants': 221, 'submit': 3909, 'https': 1860, 'innovateforhealth': 1979, 'detailed': 1074, 'description': 1058, 'list': 2294, 'requested': 3408, 'supplemental': 3947, 'materials': 2399, 'selected': 3621, 'hired': 1821, 'time': 4082, 'employee': 1295, 'starting': 3837, 'dependent': 1042, 'standards': 3832, 'nationwide': 2573, 'based': 373, 'institute': 2000, 'bakar': 359, 'sciences': 3580, 'bchsi': 381, 'travel': 4145, 'sites': 3716, 'needed': 2585, 'scientific': 3581, 'limited': 2278, 'conference': 778, 'provide': 3171, 'assistance': 290, 'relocation': 3378, 'housing': 1849, 'transportation': 4144, 'specifies': 3791, 'develops': 1091, 'executes': 1439, 'highly': 1818, 'plans': 2940, 'month': 2542, 'period': 2864, 'performs': 2863, 'directs': 1124, 'modeling': 2515, 'performance': 2857, 'integration': 2021, 'testing': 4051, 'works': 4386, 'communities': 722, 'develop': 1083, 'implement': 1911, 'optimize': 2721, 'analytics': 182, 'algorithms': 136, 'codes': 659, 'broad': 464, 'applicability': 218, 'conducts': 777, 'benchmarking': 399, 'profiling': 3126, 'tests': 4053, 'optimizes': 2723, 'workflows': 4381, 'enhance': 1332, 'capabilities': 502, 'evaluation': 1415, 'usage': 4244, 'modes': 2530, 'characteristics': 572, 'expert': 1471, 'analyses': 166, 'closely': 643, 'pharmaceuticals': 2886, 'radical': 3252, 'advancement': 91, 'identifies': 1885, 'mentors': 2442, 'scope': 3587, 'define': 1011, 'specific': 3785, 'project': 3139, 'aimed': 127, 'significantly': 3696, 'improving': 1926, 'pitches': 2930, 'proposals': 3153, 'management': 2361, 'team': 4011, 'initiates': 1975, 'involve': 2080, 'pis': 2928, 'manages': 2364, 'disclosures': 1137, 'conflict': 785, 'security': 3612, 'issues': 2100, 'publishes': 3193, 'results': 3454, 'hpc': 1852, 'techniques': 4018, 'algorithm': 134, 'enhancement': 1333, 'venues': 4282, 'promote': 3143, 'latest': 2202, 'technologies': 4024, 'community': 723, 'represent': 3399, 'organization': 2735, 'national': 2572, 'international': 2047, 'meetings': 2428, 'conferences': 779, 'committees': 714, 'contract': 826, 'propose': 3154, 'initiate': 1973, 'carry': 521, 'ambitious': 161, 'abilities': 1, 'finding': 1564, 'unmet': 4217, 'needs': 2589, 'translating': 4140, 'tractable': 4116, 'solutions': 3754, 'implemented': 1913, 'specifically': 3786, 'assigned': 283, 'self': 3626, 'motivated': 2548, 'independently': 1945, 'engage': 1321, 'variety': 4272, 'methodological': 2455, 'fields': 1546, 'mathematics': 2403, 'platforms': 2950, 'ideally': 1881, 'domains': 1191, 'medicine': 2424, 'excellent': 1434, 'regularly': 3358, 'effectively': 1261, 'communicate': 717, 'teams': 4013, 'translate': 4139, 'significant': 3695, 'effective': 1260, 'personnel': 2875, 'external': 1495, 'audiences': 321, 'resolving': 3437, 'ci': 596, 'introductory': 2068, 'intermediate': 2043, 'curate': 920, 'clean': 620, 'organize': 2739, 'messy': 2448, 'write': 4390, 'code': 657, 'unstructured': 4223, 'structured': 3890, 'programming': 3131, 'scripting': 3594, 'languages': 2195, 'python': 3207, 'java': 2112, 'scala': 3557, 'comfort': 699, 'advanced': 90, 'spark': 3772, 'addition': 70, 'hadoop': 1774, 'extraction': 1498, 'transformation': 4132, 'loading': 2307, 'etl': 1407, 'record': 3323, 'prototyping': 3168, 'scaling': 3562, 'fluent': 1586, 'environments': 1373, 'cluster': 648, 'cpus': 889, 'gpus': 1723, 'aws': 348, 'azure': 350, 'google': 1713, 'salesforce': 3539, 'ibm': 1874, 'vmware': 4323, 'extensive': 1494, 'associated': 296, 'following': 1597, 'mining': 2494, 'mathematical': 2401, 'machine': 2339, 'learning': 2224, 'deep': 1010, 'models': 2519, 'cnn': 650, 'lstm': 2328, 'gru': 1761, 'architecture': 252, 'resnet': 3433, 'net': 2596, 'frameworks': 1623, 'tensorflow': 4042, 'pytorch': 3217, 'keras': 2164, 'natural': 2574, 'language': 2193, 'processing': 3095, 'graphs': 1741, 'reinforcement': 3364, 'representation': 3400, 'optimization': 2719, 'bs': 468, 'ms': 2552, 'phd': 2889, 'applicable': 219, 'expected': 1453, 'discipline': 1132, 'physic': 2907, 'math': 2400, 'biomedical': 428, 'sectors': 3611, 'positions': 2983, 'extended': 1490, 'operational': 2712, 'demand': 1027, 'eligible': 1282, 'participate': 2806, 'welfare': 4359, 'designated': 1061, 'cover': 883, 'letter': 2250, 'cv': 937, 'single': 3711, 'attachment': 303, 'applying': 230, 'conviction': 843, 'history': 1826, 'background': 356, 'requiring': 3422, 'fingerprinting': 1569, 'check': 581, 'nature': 2576, 'hire': 1820, 'people': 2855, 'histories': 1825, 'reviews': 3484, 'received': 3300, 'context': 818, 'reserves': 3430, 'right': 3491, 'contingent': 820, 'successful': 3919, 'completion': 747, 'equal': 1379, 'affirmative': 110, 'action': 51, 'employer': 1297, 'qualified': 3224, 'consideration': 792, 'regard': 3346, 'race': 3251, 'color': 686, 'religion': 3376, 'sex': 3670, 'sexual': 3671, 'orientation': 2742, 'gender': 1678, 'identity': 1888, 'origin': 2744, 'disability': 1126, 'protected': 3159, 'veteran': 4293, 'status': 3854, 'applicant': 220, 'www': 4399, 'eeoc': 1258, 'gov': 1715, 'employers': 1298, 'upload': 4233, 'poster_screen_reader_optimized': 2996, 'pdf': 2847, 'complete': 741, 'nondiscrimination': 2631, 'policy': 2967, 'ucop': 4182, 'doc': 1175, 'nondiscrimaffirmactdemonstrated': 2628, 'postdoctoral': 2993, 'san': 3545, 'francisco': 1627, 'bay': 378, 'area': 257, 'lab': 2184, 'opening': 2704, 'scholar': 3571, 'emerging': 1287, 'novel': 2649, 'control': 835, 'applied': 227, 'infrastructure': 1968, 'networking': 2599, 'compute': 762, 'lead': 2213, 'automation': 336, 'aware': 346, 'intelligent': 2028, 'goal': 1707, 'networks': 2600, 'distributed': 1164, 'big': 414, 'department': 1038, 'energy': 1320, 'doe': 1186, 'facilities': 1506, 'developments': 1089, 'immense': 1904, 'generation': 1686, 'discoveries': 1139, 'massive': 2394, 'amounts': 164, 'movement': 2550, 'trends': 4152, 'exascale': 1430, 'leveraging': 2260, 'pattern': 2836, 'recognition': 3308, 'explore': 1482, 'quick': 3245, 'model': 2513, 'candidate': 499, 'unique': 4204, 'engaging': 1324, 'linking': 2291, 'major': 2352, 'themes': 4060, 'real': 3281, 'translation': 4141, 'impactful': 1907, 'advancing': 93, 'state': 3843, 'art': 266, 'network': 2598, 'building': 475, 'libraries': 2266, 'process': 3091, 'publicly': 3190, 'privately': 3079, 'available': 342, 'fundamental': 1652, 'properties': 3151, 'efficient': 1267, 'designs': 1067, 'innovative': 1983, 'leverage': 2258, 'publish': 3191, 'journals': 2136, 'ph': 2884, 'controls': 836, 'analytical': 180, 'foundation': 1618, 'extremely': 1499, 'aptitude': 248, 'desire': 1070, 'object': 2668, 'oriented': 2743, 'github': 1699, 'profile': 3125, 'contribute': 829, 'framework': 1622, 'demonstrate': 1032, 'scikit': 3585, 'flexible': 1580, 'matlab': 2404, 'coding': 660, 'additional': 71, 'desired': 1071, 'plus': 2956, 'deploying': 1047, 'architectures': 255, 'streaming': 3876, 'infrastructures': 1969, 'demonstration': 1034, 'posting': 2998, 'shall': 3673, 'remain': 3379, 'open': 2703, 'filled': 1556, 'appointment': 232, 'possibility': 2988, 'renewal': 3384, 'satisfactory': 3552, 'continuing': 823, 'availability': 341, 'funds': 1659, 'ongoing': 2698, 'depends': 1044, 'post': 2990, 'represented': 3401, 'union': 4203, 'collective': 680, 'bargaining': 371, 'purposes': 3199, 'predetermined': 3020, 'step': 3859, 'rates': 3270, 'subject': 3905, 'convictions': 844, 'evaluated': 1413, 'determine': 1080, 'directly': 1122, 'relate': 3365, 'requirements': 3419, 'having': 1793, 'automatically': 335, 'disqualify': 1153, 'primarily': 3066, 'performed': 2861, 'lawrence': 2206, 'cyclotron': 942, 'road': 3499, 'lbnl': 2212, 'addresses': 75, 'urgent': 4239, 'challenges': 559, 'sustainable': 3960, 'protecting': 3160, 'human': 1862, 'creating': 896, 'revealing': 3477, 'fate': 1522, 'universe': 4208, 'founded': 1619, 'expertise': 1472, 'recognized': 3311, 'nobel': 2620, 'prizes': 3081, 'office': 2683, 'rewards': 3487, 'competitive': 737, 'retirement': 3467, 'opportunities': 2715, 'view': 4299, 'click': 629, 'age': 114, 'compliance': 750, 'pay': 2840, 'transparency': 4142, 'provision': 3179, 'cfr': 550, 'poster': 2995, 'supplement': 3946, 'law': 2205, 'laboratory': 2187, 'encourages': 1314, 'women': 4374, 'minorities': 2496, 'veterans': 4294, 'groups': 1754, 'presently': 3056, 'considering': 795, 'careers': 515, 'october': 2678, 'center': 538, 'organizational': 2737, 'choir': 591, 'school': 3575, 'delivery': 1026, 'services': 3660, 'order': 2731, 'effectiveness': 1263, 'ultimately': 4187, 'accountability': 31, 'population': 2975, 'emphasize': 1293, 'innovations': 1982, 'assessment': 280, 'delivered': 1023, 'outcomes': 2749, 'cost': 866, 'involves': 2083, 'gathering': 1671, 'analyzing': 186, 'interpreting': 2056, 'managing': 2365, 'database': 958, 'processes': 3093, 'selecting': 3622, 'samples': 3542, 'preparing': 3043, 'questionnaires': 3242, 'collected': 676, 'according': 28, 'established': 1396, 'sas': 3548, 'stata': 3842, 'prepares': 3042, 'charts': 579, 'tables': 3981, 'aids': 126, 'direction': 1118, 'pi': 2914, 'pd': 2846, 'manipulates': 2369, 'purpose': 3198, 'identifying': 1887, 'internal': 2045, 'publications': 3189, 'participates': 2807, 'survey': 3957, 'instruments': 2010, 'determining': 1081, 'collaborates': 665, 'implementation': 1912, 'constituents': 801, 'assists': 294, 'business': 480, 'maintains': 2350, 'includes': 1931, 'coordinating': 850, 'investigators': 2076, 'maintain': 2347, 'administrative': 82, 'necessary': 2582, 'achieving': 41, 'examples': 1428, 'include': 1929, 'accounts': 32, 'payable': 2841, 'receivable': 3298, 'subcontract': 3904, 'manipulate': 2368, 'merge': 2445, 'perform': 2856, 'measures': 2418, 'proficiency': 3122, 'microsoft': 2474, 'suite': 3925, 'excel': 1432, 'word': 4375, 'powerpoint': 3007, 'familiarity': 1516, 'endnote': 1317, 'bibliographic': 412, 'referencing': 3341, 'organized': 2740, 'moderate': 2527, 'supervisory': 3945, 'involvement': 2082, 'initiative': 1976, 'integral': 2017, 'member': 2430, 'funding': 1656, 'ethical': 1405, 'verbally': 4284, 'written': 4395, 'masters': 2396, 'economics': 1239, 'epidemiology': 1378, 'concentration': 765, 'quantitative': 3233, 'senior': 3638, 'alumni': 156, 'relations': 3370, 'udar': 4184, 'increases': 1939, 'enhances': 1335, 'communications': 721, 'outreach': 2757, 'fundraising': 1658, 'responsible': 3449, 'centralized': 541, 'donor': 1193, 'engagement': 1322, 'strengthen': 3882, 'unit': 4205, 'offices': 2688, 'consulting': 809, 'partnering': 2815, 'fundraisers': 1657, 'handles': 1780, 'central': 540, 'encompassing': 1309, 'events': 1419, 'stewardship': 3862, 'prospect': 3156, 'gift': 1696, 'ais': 129, 'delivers': 1025, 'decentralized': 999, 'interrelated': 2058, 'cal': 486, 'cads': 484, 'activity': 58, 'encompasses': 1308, 'units': 4207, 'planning': 2939, 'strategy': 3875, 'helpdesk': 1806, 'summary': 3931, 'group': 1753, 'provides': 3176, 'set': 3663, 'analytic': 179, 'help': 1805, 'operate': 2708, 'embarks': 1285, 'campaign': 494, 'important': 1919, 'components': 755, 'shared': 3676, 'applies': 228, 'level': 2255, 'strategic': 3873, 'key': 2167, 'principal': 3068, 'planned': 2938, 'giving': 1701, 'volunteer': 4327, 'identification': 1883, 'recruitment': 3328, 'evaluating': 1414, 'predictive': 3025, 'forecasting': 1602, 'classify': 616, 'understand': 4196, 'steward': 3861, 'designing': 1066, 'implementing': 1914, 'administering': 79, 'databases': 959, 'acquire': 45, 'store': 3866, 'retrieve': 3473, 'flow': 1582, 'meet': 2426, 'requestors': 3410, 'validating': 4261, 'ensures': 1345, 'accuracy': 35, 'completeness': 744, 'establishes': 1397, 'dictionaries': 1100, 'elements': 1279, 'hypothesis': 1872, 'estimate': 1398, 'decisions': 1002, 'quantify': 3231, 'uncertainty': 4191, 'surrounding': 3956, 'decision': 1001, 'making': 2358, 'members': 2431, 'efforts': 1269, 'collaborate': 664, 'deliver': 1021, 'sophisticated': 3759, 'products': 3115, 'growing': 1756, 'schedules': 3568, 'sessions': 3662, 'synthesize': 3972, 'disseminate': 1155, 'producing': 3108, 'informative': 1965, 'presentations': 3053, 'visualizations': 4313, 'maps': 2379, 'infographics': 1962, 'continuously': 825, 'reproductibility': 3404, 'contributing': 831, 'standard': 3831, 'prepare': 3040, 'reusable': 3476, 'specifications': 3788, 'warehouses': 4337, 'import': 1917, 'stores': 3867, 'random': 3257, 'transfer': 4129, 'overall': 2761, 'patterns': 2837, 'volunteerism': 4328, 'event': 1418, 'participation': 2809, 'search': 3601, 'explanations': 1477, 'underlying': 4194, 'causes': 533, 'revenue': 3478, 'gather': 1670, 'census': 537, 'surveys': 3958, 'market': 2387, 'useful': 4247, 'share': 3675, 'stay': 3855, 'abreast': 4, 'administration': 81, 'practices': 3013, 'relating': 3368, 'logical': 2315, 'warehouse': 4336, 'web': 4346, 'able': 3, 'unassisted': 4190, 'types': 4175, 'relational': 3369, 'dimensional': 1112, 'document': 1181, 'familiar': 1515, 'common': 715, 'scholarly': 3572, 'specification': 3787, 'knowledgeable': 2178, 'governance': 1717, 'abstract': 6, 'flows': 1583, 'proven': 3170, 'attention': 308, 'interviewing': 2063, 'negotiation': 2591, 'influencing': 1961, 'oral': 2730, 'communication': 720, 'awards': 345, 'eecs': 1256, 'function': 1648, 'changing': 568, 'interprets': 2057, 'monitors': 2540, 'regarding': 3348, 'consultative': 808, 'administrators': 83, 'college': 682, 'consistently': 797, 'ranking': 3262, 'regular': 3357, 'undergraduate': 4192, 'located': 2311, 'seven': 3668, 'departments': 1040, 'largest': 2199, 'operating': 2710, 'expenditures': 1454, 'exceed': 1431, 'million': 2485, 'consisting': 798, 'adjunct': 78, 'emeriti': 1288, 'pir': 2927, 'appointments': 233, 'undergraduates': 4193, 'letters': 2253, 'evaluations': 1416, 'great': 1746, 'importance': 1918, 'reputation': 3405, 'report': 3391, 'require': 3413, 'focus': 1590, 'plan': 2937, 'submission': 3906, 'outside': 2759, 'entities': 1355, 'cra': 890, 'taulbee': 4003, 'ecedha': 1234, 'news': 2605, 'asee': 273, 'nih': 2611, 'grant': 1734, 'functions': 1651, 'advice': 102, 'committee': 713, 'chairs': 556, 'leadership': 2216, 'award': 344, 'nominations': 2624, 'sustaining': 3961, 'morale': 2547, 'person': 2870, 'nomination': 2623, 'material': 2398, 'researching': 3428, 'aspects': 276, 'processed': 3092, 'given': 1700, 'staying': 3856, 'constant': 800, 'parties': 2813, 'submissions': 3907, 'assembles': 277, 'factual': 1509, 'summarizing': 3930, 'commenting': 706, 'significance': 3694, 'matter': 2406, 'collects': 681, 'discussions': 1147, 'pertinent': 2881, 'upcoming': 4228, 'past': 2828, 'unsuccessful': 4224, 'inside': 1989, 'finalizing': 1559, 'agenda': 115, 'minutes': 2499, 'meeting': 2427, 'follows': 1598, 'participants': 2805, 'request': 3407, 'distributions': 1165, 'need': 2584, 'changes': 567, 'technological': 4023, 'archiving': 256, 'achieve': 38, 'outcome': 2748, 'announcement': 193, 'video': 4297, 'webpage': 4347, 'highlight': 1817, 'honors': 1841, 'accomplishments': 26, 'edits': 1249, 'modifies': 2531, 'documents': 1185, 'manuscripts': 2378, 'notable': 2639, 'current': 926, 'possible': 2989, 'ready': 3280, 'suggested': 3923, 'sorts': 3762, 'forms': 1611, 'completes': 745, 'sections': 3609, 'submitted': 3911, 'ap': 207, 'bears': 384, 'inspec': 1994, 'patent': 2829, 'trademark': 4117, 'provided': 3172, 'stays': 3857, 'contact': 812, 'completed': 742, 'timely': 4087, 'lunch': 2333, 'dinner': 1113, 'distinguished': 1162, 'breakfast': 455, 'logistics': 2317, 'reception': 3305, 'winners': 4370, 'writes': 4392, 'scripts': 3595, 'guide': 1765, 'broader': 465, 'drafts': 1203, 'follow': 1596, 'summaries': 3927, 'addendums': 68, 'speeches': 3795, 'briefings': 459, 'executive': 1440, 'correspondence': 864, 'roles': 3511, 'appropriate': 240, 'managers': 2363, 'supporting': 3951, 'updates': 4231, 'enhancements': 1334, 'finer': 1567, 'distinctions': 1160, 'resource': 3439, 'director': 1123, 'compiles': 740, 'behalf': 390, 'contacts': 815, 'internally': 2046, 'exchange': 1436, 'demanding': 1028, 'timeframes': 4084, 'highest': 1816, 'discretion': 1142, 'strict': 3885, 'confidentiality': 782, 'sound': 3763, 'judgment': 2146, 'responding': 3443, 'concerns': 770, 'analyze': 184, 'accurately': 37, 'synthesizing': 3973, 'bachelors': 354, 'tuesday': 4165, 'oct': 2677, 'pm': 2962, 'pacific': 2779, 'reviewed': 3481, 'final': 1557, 'apr': 245, 'continue': 821, 'accepted': 18, 'seeks': 3616, 'scholars': 3573, 'teach': 4006, 'build': 474, 'datascience': 972, 'number': 2655, 'varies': 4271, 'semester': 3630, 'depending': 1043, 'typically': 4176, 'start': 3835, 'january': 2111, 'august': 326, 'renewed': 3385, 'online': 2699, 'participating': 2808, 'pursuing': 3201, 'directed': 1117, 'presenting': 3055, 'lecturing': 2242, 'holding': 1834, 'hours': 1846, 'grading': 1729, 'assigning': 284, 'grades': 1727, 'advising': 105, 'course': 878, 'postdocs': 2992, 'small': 3736, 'lecturer': 2239, 'newest': 2603, 'created': 895, 'address': 74, 'society': 3743, 'sense': 3641, 'abundance': 7, 'collect': 675, 'distance': 1158, 'way': 4342, 'govern': 1716, 'shape': 3674, 'private': 3078, 'goals': 1708, 'educates': 1251, 'possibilities': 2987, 'practice': 3012, 'usable': 4243, 'offer': 2680, 'degrees': 1018, 'mims': 2487, 'trains': 4123, 'emphasizes': 1294, 'classes': 611, 'mics': 2476, 'cybersecurity': 940, 'leaders': 2215, 'contextual': 819, 'mids': 2481, 'completely': 743, 'train': 4120, 'equips': 1383, 'influence': 1960, 'sharing': 3678, 'impressive': 1920, 'known': 2180, 'tech': 4016, 'companies': 724, 'nonprofits': 2632, 'sector': 3610, 'advantage': 94, 'ground': 1752, 'floor': 1581, 'create': 894, 'ups': 4236, 'basic': 375, 'enrolled': 1340, 'earned': 1228, 'sociology': 3744, 'postdoc': 2991, 'core': 857, 'areas': 258, 'analysisapplied': 168, 'statisticsdata': 3852, 'storage': 3865, 'retrievalvisualizing': 3472, 'communicating': 719, 'dataapplied': 956, 'learningprivacy': 2234, 'ethics': 1406, 'dataexperiments': 966, 'causal': 532, 'inferencevery': 1959, 'regression': 3355, 'series': 3652, 'analysisdeep': 171, 'learningsocial': 2235, 'analysisnatural': 175, 'authored': 328, 'articles': 268, 'broadly': 466, 'defined': 1012, 'limits': 2279, 'assignments': 286, 'interact': 2030, 'multidisciplinary': 2556, 'environment': 1362, 'salaries': 3536, 'total': 4110, 'duration': 1219, 'individual': 1950, 'link': 2289, 'aprecruit': 246, 'jpf': 2138, 'pool': 2971, 'april': 247, 'accommodate': 22, 'starts': 3838, 'fall': 1514, 'spring': 3807, 'summer': 3932, 'wish': 4372, 'reapply': 3290, 'reference': 3339, 'treated': 4147, 'confidential': 781, 'arrange': 262, 'recommendation': 3314, 'uploaded': 4234, 'recommenders': 3317, 'refer': 3336, 'referees': 3338, 'party': 2820, 'dossier': 1194, 'career': 514, 'statement': 3844, 'apo': 213, 'evalltr': 1411, 'submitting': 3912, 'nondiscrimaffirmact': 2627, 'interested': 2036, 'diversity': 1168, 'addressing': 76, 'contributions': 832, 'employees': 1296, 'balance': 362, 'direct': 1116, 'questions': 3243, 'dean': 994, 'ischool': 2094, 'location': 2312, 'ischooljobs': 2095, 'datascipostdoc': 973, 'curriculum': 928, 'vitae': 4321, 'recently': 3304, 'updated': 4230, 'interests': 2038, 'approach': 237, 'indicate': 1948, 'class': 610, 'es': 1389, 'equity': 1385, 'inclusion': 1933, 'topics': 4109, 'ofew': 2679, 'copy': 854, 'recent': 3303, 'publication': 3188, 'manuscript': 2377, 'optional': 2726, 'encourage': 1311, 'send': 3636, 'taught': 4002, 'instructor': 2008, 'assistant': 291, 'requiredresearch': 3418, 'processingcover': 3098, 'lettercurriculum': 2252, 'approachstatement': 239, 'preparationpdf': 3039, 'requir': 3412, 'geochemistry': 1692, 'earth': 1229, 'environmental': 1364, 'eesa': 1259, 'join': 2126, 'theories': 4061, 'reactive': 3276, 'multicomponent': 2555, 'subsurface': 3915, 'role': 3505, 'molecular': 2534, 'modelers': 2514, 'extract': 1496, 'geological': 1693, 'synthetic': 3974, 'generated': 1684, 'simulations': 3709, 'component': 754, 'aqueous': 250, 'mineral': 2491, 'solution': 3752, 'interfaces': 2041, 'atmospheric': 302, 'frequently': 1633, 'extremes': 1500, 'temperature': 4036, 'pressure': 3059, 'challenge': 557, 'predict': 3021, 'chemical': 584, 'thermodynamics': 4065, 'kinetics': 2172, 'fluids': 1588, 'inadequate': 1927, 'limit': 2276, 'water': 4341, 'driving': 1211, 'paradigm': 2794, 'shift': 3682, 'ranging': 3259, 'image': 1898, 'speech': 3794, 'drug': 1212, 'test': 4049, 'simulation': 3707, 'correlations': 863, 'responses': 3446, 'relationships': 3371, 'feature': 1525, 'chemistry': 585, 'physics': 2913, 'layers': 2209, 'constraints': 802, 'interpretable': 2055, 'ml': 2505, 'geoscience': 1694, 'present': 3051, 'peer': 2852, 'physical': 2908, 'caffe': 485, 'nwchem': 2662, 'dl_poly': 1174, 'lammps': 2190, 'gromacs': 1750, 'vasp': 4275, 'cp': 885, 'overtime': 2768, 'global': 1704, 'cega': 534, 'premiere': 3037, 'affiliates': 109, 'powerful': 3006, 'rigorous': 3493, 'measure': 2415, 'impacts': 1909, 'sampling': 3543, 'assisting': 293, 'boards': 435, 'input': 1984, 'funders': 1655, 'papers': 2793, 'literature': 2298, 'drafting': 1202, 'portions': 2981, 'organizing': 2741, 'run': 3526, 'cleaning': 621, 'accessible': 20, 'repositories': 3396, 'summarizes': 3929, 'analysts': 178, 'microeconomics': 2469, 'econometrics': 1237, 'consultation': 806, 'hands': 1782, 'handle': 1779, 'different': 1105, 'tasks': 4000, 'deadlines': 992, 'hourly': 1845, 'careerspub': 517, 'psp': 3180, 'ucb': 4179, 'hrms': 1854, 'hrs_hram': 1857, 'hrs_app_schjob': 1856, 'gbl': 1673, 'page': 2783, 'hrs_app_jbpst': 1855, 'siteid': 3715, 'jobopeningid': 2123, 'postingseq': 2999, 'mar': 2380, 'tenure': 4045, 'track': 4113, 'lecturers': 2240, 'courses': 879, 'screening': 3592, 'duties': 1220, 'minimum': 2493, 'qualification': 3221, 'enrollment': 1341, 'coursesan': 880, 'programmingresearch': 3132, 'fiscal': 1573, 'generally': 1682, 'hires': 1822, 'basis': 376, 'march': 2381, 'semesters': 3631, 'note': 2641, 'guarantee': 1762, 'exists': 1448, 'specified': 3790, 'currently': 927, 'reviewing': 3483, 'future': 1661, 'solicited': 3748, 'finalists': 1558, 'datasci': 971, 'interlibrary': 2042, 'ils': 1897, 'library': 2267, 'borrowing': 447, 'lending': 2247, 'clientele': 633, 'independent': 1944, 'comprised': 759, 'photoduplication': 2904, 'ill': 1894, 'requests': 3411, 'clients': 634, 'worldwide': 4388, 'loan': 2308, 'duplication': 1218, 'held': 1804, 'agreement': 120, 'sal': 3535, 'section': 3608, 'campuses': 496, 'stanford': 3834, 'ut': 4252, 'austin': 327, 'ibs': 1875, 'patrons': 2835, 'wishing': 4373, 'obtain': 2672, 'baker': 360, 'fee': 1529, 'used': 4246, 'specialized': 3780, 'equipment': 1382, 'payments': 2842, 'owed': 2771, 'iii': 1892, 'supervisor': 3944, 'digitization': 1111, 'partnership': 2817, 'colleagues': 674, 'afield': 113, 'ls': 2327, 'approximately': 243, 'sle': 3731, 'guidelines': 1767, 'serve': 3653, 'leader': 2214, 'monitoring': 2539, 'sles': 3733, 'sending': 3637, 'stacks': 3822, 'inspecting': 1995, 'scan': 3563, 'annotating': 190, 'picklist': 2916, 'spreadsheet': 3806, 'recording': 3325, 'reasons': 3295, 'volumes': 4326, 'scanning': 3566, 'book': 441, 'carts': 523, 'lot': 2325, 'numbers': 2656, 'scheme': 3570, 'outgoing': 2750, 'incoming': 1935, 'dock': 1176, 'staging': 3828, 'shelving': 3681, 'items': 2101, 'return': 3474, 'coordinate': 848, 'difficult': 1107, 'arising': 261, 'workflow': 4380, 'recognizing': 3312, 'forward': 1615, 'head': 1797, 'resolution': 3434, 'maintaining': 2349, 'procedural': 3087, 'instructional': 2007, 'specialty': 3783, 'mail': 2344, 'room': 3514, 'ii': 1891, 'manifest': 2367, 'reconciliation': 3320, 'errors': 1388, 'mismatched': 2500, 'barcodes': 369, 'compile': 738, 'daily': 946, 'updating': 4232, 'cumulative': 919, 'iv': 2107, 'consult': 804, 'transport': 4143, 'preservation': 3058, 'telephone': 4032, 'calls': 490, 'discuss': 1145, 'checking': 582, 'collections': 679, 'bancroft': 365, 'oskicat': 2747, 'millennium': 2484, 'melvyl': 2429, 'article': 267, 'engines': 1330, 'resolves': 3436, 'vague': 4259, 'incomplete': 1936, 'citations': 601, 'english': 1331, 'routes': 3518, 'fulfillment': 1645, 'update': 4229, 'unfilled': 4202, 'vdx': 4277, 'oclc': 2676, 'clio': 640, 'apprise': 235, 'scanned': 3564, 'drives': 1210, 'uploads': 4235, 'electronically': 1277, 'returnables': 3475, 'utilize': 4253, 'overhead': 2764, 'flatbed': 1579, 'scanners': 3565, 'microfilm': 2470, 'readers': 3278, 'absence': 5, 'workers': 4379, 'recognize': 3310, 'matching': 2397, 'operation': 2711, 'involving': 2084, 'phone': 2898, 'email': 1283, 'execute': 1438, 'simultaneously': 3710, 'lift': 2273, 'pounds': 3002, 'push': 3202, 'pull': 3195, 'loaded': 2306, 'weighing': 4354, 'succeed': 3916, 'fluctuations': 1584, 'differing': 1106, 'styles': 3903, 'circulation': 598, 'proceduresexperience': 3090, 'volume': 4325, 'fast': 1520, 'paced': 2778, 'layout': 2210, 'hour': 1844, 'eligibility': 1281, 'resume': 3460, 'nondiscrimaffirmactsending': 2630, 'espm': 1393, 'cnr': 652, 'biological': 425, 'production': 3110, 'extension': 1493, 'divisions': 1171, 'ecosystem': 1241, 'organisms': 2734, 'budget': 471, 'federal': 1528, 'cooperative': 847, 'specialists': 3777, 'fieldwork': 1551, 'consists': 799, 'commercial': 708, 'collecting': 677, 'plant': 2942, 'counting': 874, 'insects': 1988, 'insectary': 1987, 'colonies': 685, 'enemies': 1318, 'microscope': 2471, 'sort': 3761, 'count': 873, 'insect': 1986, 'biology': 427, 'ecology': 1235, 'entry': 1360, 'quarantine': 3235, 'helping': 1808, 'safety': 3531, 'protocols': 3163, 'instructing': 2005, 'structure': 3889, 'cards': 512, 'rearing': 3292, 'diet': 1103, 'feeding': 1531, 'spiders': 3797, 'black': 431, 'widow': 4365, 'species': 3784, 'proper': 3150, 'supervision': 3943, 'doctorate': 1180, 'willing': 4367, 'hot': 1843, 'cool': 846, 'weather': 4345, 'attentive': 309, 'details': 1075, 'focuses': 1592, 'agriculture': 122, 'pests': 2883, 'entomology': 1357, 'personality': 2872, 'alongside': 151, 'near': 2579, 'pesticides': 2882, 'competent': 733, 'safe': 3529, 'eye': 1501, 'hand': 1778, 'dexterity': 1094, 'docs': 1178, 'internet': 2048, 'weekly': 4353, 'friday': 1637, 'nov': 2648, 'capstone': 509, 'invites': 2079, 'temporary': 4037, 'instructors': 2009, 'arise': 260, 'stat': 3841, 'introduction': 2067, 'probability': 3083, 'concepts': 767, 'stochastic': 3863, 'linear': 2282, 'theory': 4062, 'modern': 2529, 'prediction': 3022, 'game': 1667, 'seminar': 3633, 'reproducible': 3403, 'collaborative': 670, 'syllabi': 3966, 'exams': 1429, 'supervising': 3942, 'assistants': 292, 'chair': 553, 'documented': 1183, 'credit': 900, 'session': 3661, 'philosophyteaching': 2897, 'names': 2568, 'solicit': 3747, 'tenured': 4046, 'lettercopy': 2251, 'transcript': 4126, 'november': 2650, 'close': 642, 'philosophy': 2895, 'letterstatement': 2254, 'philosophystatement': 2896, 'evaluationscopy': 1417, 'trilogy': 4154, 'partners': 2816, 'universities': 4209, 'ux': 4256, 'ui': 4185, 'platform': 2948, 'combines': 695, 'robust': 3503, 'multinational': 2559, 'adult': 87, 'learners': 2222, 'digital': 1110, 'economy': 1240, 'passion': 2823, 'individuals': 1952, 'fresh': 1634, 'seek': 3614, 'transition': 4137, 'classroom': 618, 'inquiry': 1985, 'meta': 2450, 'feedback': 1530, 'don': 1192, 'ton': 4099, 'just': 2154, 'lifelong': 2271, 'learner': 2221, 'teacher': 4007, 'love': 2326, 'change': 566, 'lives': 2302, 'attendance': 305, 'bootcampspot': 445, 'regards': 3349, 'walk': 4331, 'unable': 4188, 'grade': 1725, 'homework': 1840, 'fit': 1574, 'proficient': 3123, 'pandas': 2789, 'matplotlib': 2405, 'beautiful': 386, 'soup': 3764, 'javascript': 2113, 'css': 913, 'leaflet': 2218, 'nosql': 2637, 'bootcampspotensure': 446, 'timewalk': 4091, 'studentsresearch': 3895, 'tograde': 4097, 'assignmentspython': 288, 'pandasmatplotlibbeautiful': 2790, 'soupjavascripthtml': 3766, 'leafletsql': 2219, 'nosqltableaumachine': 2638, 'pennsylvania': 2854, 'title': 4093, 'pca': 2844, 'type': 4174, 'churchill': 595, 'posted': 2994, 'expires': 1474, 'pace': 2777, 'esy': 1402, 'behavioral': 392, 'therapeutic': 4063, 'diploma': 1114, 'ged': 1677, 'checks': 583, 'pa': 2776, 'dept': 1051, 'tb': 4004, 'pre': 3016, 'essential': 1394, 'clerical': 627, 'instruction': 2006, 'behavior': 391, 'coordinator': 851, 'objectives': 2669, 'identified': 1884, 'iep': 1890, 'revision': 3485, 'indicated': 1949, 'functional': 1649, 'orderly': 2732, 'behaviors': 394, 'procedure': 3088, 'logs': 2318, 'interventions': 2061, 'certification': 548, 'adopted': 85, 'crisis': 906, 'intervention': 2060, 'verbal': 4283, 'escalation': 1390, 'nonconfrontational': 2626, 'setting': 3665, 'restraint': 3451, 'utilizes': 4254, 'forth': 1614, 'accordance': 27, 'resort': 3438, 'harming': 1788, 'arrival': 265, 'dismissal': 1150, 'transitions': 4138, 'supervises': 3941, 'day': 985, 'schedule': 3567, 'unless': 4214, 'officially': 2689, 'professor': 3120, 'th': 4057, 'july': 2149, 'st': 3817, 'regarded': 3347, 'values': 4269, 'fully': 1646, 'realized': 3285, 'commitment': 710, 'evidence': 1420, 'granting': 1735, 'preferably': 3027, 'finish': 1570, 'promoting': 3144, 'profession': 3116, 'writers': 4391, 'url': 4240, 'sure': 3953, 'notice': 2646, 'dual': 1217, 'couples': 877, 'parents': 2802, 'preprints': 3044, 'thesis': 4066, 'requiredcurriculum': 3415, 'intereststatement': 2039, 'researchstatement': 3429, 'teachingstatement': 4010, 'associate': 295, 'dec': 997, 'hold': 1831, 'institution': 2002, 'notify': 2647, 'contacting': 814, 'december': 998, 'hr': 1853, 'preeminent': 3026, 'centers': 539, 'places': 2936, 'emphasis': 1292, 'partnerships': 2818, 'sloan': 3735, 'simons': 3703, 'citadel': 600, 'street': 3880, 'genentech': 1679, 'bioengineering': 422, 'biostatistics': 429, 'electrical': 1275, 'demography': 1029, 'ib': 1873, 'integrative': 2022, 'mcb': 2409, 'cell': 536, 'emeritus': 1289, 'dizzying': 1172, 'array': 263, 'medal': 2420, 'copss': 853, 'academy': 12, 'attractive': 316, 'visiting': 4307, 'manage': 2359, 'preceding': 3017, 'case': 524, 'coming': 701, 'reconcile': 3319, 'staffing': 3824, 'sent': 3643, 'actions': 53, 'cases': 525, 'ladder': 2189, 'rank': 3260, 'managed': 2360, 'appraisal': 234, 'merit': 2447, 'sabbatical': 3528, 'leaves': 2237, 'recalls': 3297, 'chairholder': 554, 'separation': 3644, 'initiated': 1974, 'fashion': 1519, 'draft': 1201, 'memos': 2436, 'assure': 300, 'prepared': 3041, 'accurate': 36, 'promotion': 3145, 'advise': 103, 'credits': 901, 'buy': 482, 'outs': 2758, 'administers': 80, 'phases': 2888, 'selection': 3623, 'soliciting': 3749, 'estimates': 1400, 'advertising': 101, 'ads': 86, 'submits': 3910, 'advertisements': 99, 'line': 2281, 'coordinates': 849, 'reviewers': 3482, 'concerning': 769, 'smooth': 3740, 'recruitments': 3329, 'reappointments': 3291, 'promotions': 3147, 'serves': 3656, 'point': 2965, 'visitor': 4308, 'regional': 3351, 'partner': 2814, 'titles': 4094, 'generalist': 1681, 'correct': 861, 'da': 944, 'approval': 242, 'overseeing': 2766, 'timekeeping': 4085, 'payroll': 2843, 'corrections': 862, 'agreements': 121, 'counsel': 869, 'manager': 2362, 'advises': 104, 'champs': 564, 'helps': 1809, 'counsels': 872, 'leave': 2236, 'counseling': 870, 'juxtaposes': 2156, 'historical': 1824, 'pertaining': 2880, 'consults': 810, 'visa': 4301, 'pr': 3009, 'considerations': 793, 'assignment': 285, 'files': 1555, 'retreats': 3470, 'commencement': 703, 'bstars': 470, 'attends': 307, 'votes': 4329, 'newsletter': 2606, 'listservs': 2297, 'pages': 2784, 'responds': 3444, 'recommends': 3318, 'comply': 753, 'interpret': 2053, 'nuanced': 2652, 'thorough': 4072, 'contracts': 828, 'priorities': 3073, 'legal': 2244, 'implications': 1916, 'officials': 2690, 'tact': 3984, 'diplomacy': 1115, 'hcm': 1796, 'apbears': 209, 'appointee': 231, 'pedagogical': 2849, 'secondary': 3606, 'achievement': 39, 'pedagogy': 2851, 'pedagogically': 2850, 'industrial': 1953, 'engineer': 1326, 'jquery': 2140, 'node': 2621, 'js': 2143, 'react': 3273, 'git': 1697, 'command': 702, 'mysql': 2564, 'php': 2905, 'positive': 2984, 'attitude': 310, 'bonuses': 440, 'ruby': 3524, 'rails': 3254, 'sequelize': 3646, 'express': 1487, 'assignmentsjavascripthtml': 287, 'jquerynode': 2141, 'jsreact': 2145, 'jsgitcommand': 2144, 'linemysqlphp': 2284, 'frameworksdemonstrated': 1624, 'settinga': 3666, 'attituderuby': 312, 'railssequelizeexpresspython': 3255, 'souphtml': 3765, 'seattle': 3604, 'wa': 4330, 'siri': 3713, 'apple': 215, 'customers': 932, 'dialects': 1097, 'challenging': 562, 'deploy': 1045, 'vision': 4303, 'lingual': 2286, 'macos': 2342, 'ios': 2087, 'sdks': 3596, 'millions': 2486, 'nlp': 2614, 'similar': 3702, 'shipping': 3685, 'mainstream': 2346, 'like': 2275, 'weka': 4356, 'nltk': 2615, 'et': 1403, 'al': 131, 'practical': 3010, 'quickly': 3246, 'prototype': 3165, 'ideas': 1882, 'solving': 3757, 'cross': 909, 'looking': 2322, 'technologist': 4029, 'problem': 3084, 'ship': 3684, 'surprise': 3955, 'delight': 1020, 'directions': 1120, 'specialization': 3778, 'teamsclear': 4014, 'commun': 716, 'serviceable': 3658, 'condition': 773, 'takes': 3987, 'disaster': 1128, 'response': 3445, 'salvage': 3540, 'binding': 420, 'repair': 3387, 'conservation': 790, 'treatment': 4148, 'replacement': 3389, 'damaged': 949, 'deteriorated': 1079, 'repaired': 3388, 'enable': 1306, 'continued': 822, 'orders': 2733, 'mend': 2437, 'circulating': 597, 'bindery': 419, 'monographs': 2541, 'serials': 3651, 'spine': 3799, 'marking': 2392, 'cloth': 646, 'bound': 448, 'requisition': 3423, 'pack': 2780, 'trucks': 4157, 'plastic': 2947, 'lugs': 2332, 'route': 3517, 'books': 442, 'stations': 3849, 'charged': 575, 'rebinding': 3296, 'casing': 526, 'discharge': 1130, 'liaison': 2264, 'paper': 2792, 'reformatting': 3343, 'supervise': 3939, 'photocopying': 2903, 'print': 3070, 'missing': 2501, 'owned': 2772, 'collation': 673, 'illegible': 1895, 'text': 4055, 'brittle': 463, 'inspection': 1996, 'photocopies': 2901, 'prd': 3015, 'legibility': 2246, 'margins': 2383, 'evaluate': 1412, 'lapses': 2197, 'routing': 3521, 'copying': 855, 'sheet': 3679, 'copies': 852, 'editions': 1248, 'reprints': 3402, 'formats': 1609, 'originals': 2746, 'verify': 4286, 'photocopy': 2902, 'searching': 3602, 'locate': 2310, 'source': 3767, 'printouts': 3071, 'bib': 411, 'issue': 2099, 'feasible': 1524, 'costs': 867, 'copyright': 856, 'clearance': 625, 'permanent': 2867, 'file': 1554, 'permissions': 2868, 'denied': 1035, 'selectors': 3625, 'format': 1608, 'expense': 1455, 'unanticipated': 4189, 'choices': 590, 'allocate': 149, 'suggest': 3922, 'operations': 2713, 'sign': 3691, 'specify': 3792, 'treatments': 4149, 'needing': 2588, 'commitments': 711, 'responsive': 3450, 'space': 3771, 'supplies': 3948, 'attend': 304, 'pres': 3048, 'independence': 1943, 'competence': 732, 'nondiscrimaffirmactprocess': 2629, 'week': 4351, 'mon': 2535, 'fri': 1635, 'break': 454, 'programmon': 3133, 'friclass': 1636, 'scientist': 3582, 'cambridge': 492, 'ma': 2337, 'feb': 1527, 'ideal': 1880, 'taking': 3988, 'conversational': 838, 'agents': 116, 'investigate': 2073, 'customer': 930, 'experiment': 1465, 'play': 2951, 'defining': 1014, 'crossroads': 910, 'bues': 472, 'hyannis': 1868, 'caring': 520, 'reinforce': 3363, 'initially': 1972, 'introduced': 2064, 'children': 589, 'early': 1226, 'childhood': 588, 'elementary': 1278, 'helpful': 1807, 'alternative': 153, 'acceptable': 16, 'hinder': 1819, 'physically': 2909, 'challenged': 558, 'cpi': 886, 'cpr': 887, 'aid': 125, 'toileting': 4098, 'hygiene': 1869, 'awareness': 347, 'allergies': 147, 'seizures': 3619, 'medication': 2423, 'protocol': 3162, 'adapt': 63, 'easily': 1230, 'unpredictable': 4221, 'accompany': 24, 'enrichments': 1339, 'specials': 3782, 'recess': 3306, 'modify': 2532, 'academics': 11, 'accordingly': 29, 'quizzes': 3248, 'consultations': 807, 'teachers': 4008, 'providers': 3175, 'cst': 914, 'desk': 1072, 'pupil': 3196, 'additonal': 73, 'notation': 2640, 'tlc': 4095, 'sail': 3534, 'sed': 3613, 'terms': 4048, 'talented': 3990, 'realtime': 3289, 'relevance': 3373, 'discover': 1138, 'hidden': 1812, 'vast': 4276, 'smarter': 3738, 'better': 408, 'doing': 1188, 'integrated': 2019, 'nn': 2618, 'naive': 2566, 'bayes': 379, 'svm': 3962, 'forests': 1606, 'toolkits': 4101, 'numpy': 2659, 'visualisation': 4311, 'ggplot': 1695, 'hive': 1827, 'pig': 2918, 'mongodb': 2536, 'cassandra': 527, 'hbase': 1794, 'realm': 3287, 'android': 188, 'studio': 3898, 'overwhelming': 2770, 'weight': 4355, 'desirablegreat': 1069, 'skillsexperience': 3726, 'pigexperience': 2919, 'realmgood': 3288, 'studiodata': 3899, 'researcher': 3425, 'targeted': 3997, 'tbr': 4005, 'lines': 2285, 'informs': 1967, 'tennessee': 4040, 'stakeholders': 3829, 'actionable': 52, 'assess': 278, 'cycle': 941, 'benefit': 400, 'target': 3996, 'personalized': 2873, 'user': 4248, 'ended': 1315, 'merges': 2446, 'timelines': 4086, 'turn': 4168, 'conducting': 776, 'willingness': 4368, 'short': 3688, 'long': 2319, 'term': 4047, 'courteous': 882, 'trustworthy': 4160, 'player': 2952, 'coursework': 881, 'tenets': 4039, 'behaviorally': 393, 'informed': 1966, 'obtained': 2673, 'closing': 644, 'req': 3406, 'ev': 1410, 'empower': 1301, 'features': 1526, 'classifiers': 615, 'extending': 1491, 'company': 725, 'enhancing': 1336, 'cleansing': 623, 'verifying': 4287, 'integrity': 2023, 'automated': 334, 'anomaly': 198, 'detection': 1078, 'spotfire': 3805, 'sap': 3547, 'speak': 3774, 'plebeian': 2953, 'sqoop': 3813, 'light': 2274, 'believe': 395, 'evtechnologies': 1422, 'com': 690, 'foothill': 1599, 'ranch': 3256, 'tae': 3985, 'tri': 4153, 'alpha': 152, 'sized': 3718, 'beta': 407, 'magnetically': 2343, 'confined': 784, 'fusion': 1660, 'plasmas': 2946, 'beam': 382, 'reversed': 3479, 'configuration': 783, 'built': 477, 'aka': 130, 'norman': 2635, 'plasma': 2945, 'experiments': 1468, 'parameter': 2799, 'regimes': 3350, 'power': 3003, 'plants': 2944, 'diagnostic': 1095, 'fertile': 1540, 'biggest': 415, 'entire': 1354, 'pipeline': 2922, 'acquisition': 47, 'distil': 1159, 'raw': 3271, 'iterating': 2105, 'convert': 841, 'routines': 3520, 'stages': 3826, 'product': 3109, 'debugging': 996, 'maintenance': 2351, 'collaboratively': 671, 'scipy': 3586, 'professionally': 3118, 'exposure': 1486, 'bayesian': 380, 'inference': 1957, 'experimental': 1466, 'linehelp': 2283, 'pythonexperience': 3213, 'levelspassion': 2257, 'kerasexposure': 2165, 'datafamiliarity': 968, 'lin': 2280, 'intern': 2044, 'diego': 1102, 'powers': 3008, 'exactly': 1424, 'lytx': 2336, 'hungry': 1867, 'humble': 1865, 'capable': 504, 'difference': 1104, 'save': 3554, 'internship': 2050, 'monitor': 2538, 'drivers': 1209, 'risk': 3494, 'wants': 4335, 'gain': 1663, 'solve': 3755, 'client': 632, 'retention': 3466, 'satisfaction': 3550, 'compare': 728, 'led': 2243, 'deployed': 1046, 'accredited': 33, 'conventional': 837, 'cnns': 651, 'rnns': 3498, 'transformer': 4134, 'proud': 3169, 'free': 1629, 'attracting': 315, 'retaining': 3464, 'maximizing': 2408, 'inclusive': 1934, 'workforce': 4382, 'eoe': 1376, 'disabled': 1127, 'vet': 4292, 'satisfactionanalyze': 3551, 'performanceassist': 2858, 'backgroundworking': 358, 'plusexperience': 2958, 'preferredbasic': 3029, 'databas': 957, 'dissemination': 1156, 'sleep': 3732, 'advertisement': 98, 'aharvey': 123, 'psychology': 3183, 'golden': 1710, 'bear': 383, 'mood': 2546, 'centre': 542, 'gbsmrc': 1674, 'allison': 148, 'harvey': 1790, 'adding': 69, 'talent': 3989, 'come': 697, 'nursing': 2660, 'funded': 1654, 'nimh': 2612, 'nichd': 2610, 'theme': 4059, 'psychosocial': 3185, 'psychiatric': 3181, 'harnessing': 1789, 'mental': 2438, 'transdiagnostic': 4128, 'severe': 3669, 'illness': 1896, 'memory': 2435, 'depression': 1050, 'studying': 3901, 'parent': 2801, 'teen': 4031, 'conversations': 839, 'mentored': 2440, 'collaborations': 669, 'existing': 1447, 'dr': 1200, 'licensed': 2268, 'psychologist': 3182, 'potentially': 3001, 'incumbent': 1942, 'dissertation': 1157, 'proof': 3148, 'conferral': 780, 'said': 3533, 'asset': 281, 'settings': 3667, 'local': 2309, 'anticipated': 205, 'begin': 389, 'soon': 3758, 'negotiable': 2590, 'commit': 709, 'dental': 1036, 'coverage': 884, 'included': 1930, 'going': 1709, 'describing': 1057, 'discussion': 1146, 'qualities': 3226, 'listed': 2295, 'particular': 2811, 'outline': 2754, 'specifics': 3789, 'esp': 1391, 'hierarchical': 1813, 'multilevel': 2557, 'belmont': 398, 'existence': 1446, 've': 4278, 'launched': 2204, 'nation': 2571, 'piece': 2917, 'infuse': 1970, 'empathy': 1291, 'encouragement': 1313, 'fun': 1647, 'lesson': 2248, 'country': 875, 'experiences': 1463, 'insight': 1990, 'isn': 2096, 'drive': 1207, 'growth': 1759, 'api': 210, 'interactions': 2032, 'popular': 2974, 'passionate': 2824, 'empowering': 1302, 'hear': 1801, 'transformative': 4133, 'path': 2830, 'rewarding': 3486, 'prerequisite': 3047, 'll': 2303, 'journey': 2137, 'impacting': 1908, 'lectures': 2241, 'panopto': 2791, 'turned': 4169, 'actively': 55, 'lecture': 2238, 'repository': 3397, 'graded': 1726, 'occasionally': 2675, 'classrooms': 619, 'body': 436, 'individually': 1951, 'emotional': 1290, 'roller': 3512, 'coaster': 655, 'wed': 4350, 'sat': 3549, 'tue': 4164, 'thu': 4078, 'weekday': 4352, 'saturday': 3553, 'zr': 4413, 'activitiesanswer': 57, 'stagewalk': 3827, 'neededensure': 2587, 'improveupload': 1925, 'repositorycreate': 3398, 'gradingpython': 1730, 'hadoopbachelor': 1775, 'degreeminimum': 1017, 'experiencea': 1457, 'attitudeability': 311, 'bodyshare': 437, 'studentssupport': 3896, 'coasterbe': 656, 'scheduleweekday': 3569, 'york': 4408, 'ny': 2666, 'thought': 4074, 'tackling': 3983, 'smile': 3739, 'got': 1714, 'knack': 2175, 'dollar': 1189, 'answered': 200, 'yes': 4407, 'produces': 3107, 'prescriptive': 3049, 'marketing': 2388, 'segmentation': 3618, 'sales': 3538, 'gained': 1664, 'adept': 77, 'running': 3527, 'comfortable': 700, 'discovering': 1140, 'makers': 2354, 'valuable': 4265, 'automate': 333, 'undertake': 4201, 'preprocessing': 3046, 'draw': 1204, 'clustering': 649, 'tree': 4150, 'neural': 2601, 'advantages': 95, 'drawbacks': 1205, 'glm': 1702, 'forest': 1605, 'boosting': 443, 'trees': 4151, 'powerbi': 3004, 'leading': 2217, 'consumer': 811, 'finance': 1561, 'specializes': 3781, 'card': 511, 'debt': 995, 'named': 2567, 'best': 406, 'certified': 549, 'times': 4090, 'fastest': 1521, 'city': 606, 'prestigious': 3060, 'analysiscommunicate': 170, 'makersidentify': 2355, 'processesundertake': 3094, 'dataanalyze': 955, 'patternsbuild': 2838, 'algorithmspresent': 141, 'techniquescollaborate': 4019, 'experiencegraduate': 1459, 'redwood': 3335, 'priscilla': 3076, 'chan': 565, 'mark': 2386, 'zuckerberg': 4414, 'czi': 943, 'kind': 2170, 'philanthropy': 2894, 'toughest': 4112, 'eradicating': 1387, 'disease': 1148, 'reforming': 3344, 'criminal': 905, 'justice': 2155, 'pairing': 2786, 'investing': 2077, 'healthy': 1800, 'designers': 1063, 'facing': 1507, 'ensuring': 1346, 'lucky': 2331, 'tailored': 3986, 'aspect': 275, 'grantmaking': 1736, 'advocacy': 108, 'young': 4409, 'enter': 1349, 'adulthood': 88, 'equip': 1381, 'summit': 3933, 'habits': 1771, 'wellbeing': 4360, 'measuring': 2419, 'cognitive': 662, 'onboard': 2695, 'onboarding': 2696, 'defines': 1013, 'measurements': 2417, 'metrics': 2464, 'communicates': 718, 'users': 4249, 'pilots': 2921, 'educators': 1254, 'inform': 1963, 'roadmap': 3500, 'insightful': 1991, 'dashboards': 951, 'exploratory': 1481, 'dive': 1166, 'varying': 4274, 'sophistication': 3760, 'experimentation': 1467, 'psychometrics': 3184, 'exploration': 1480, 'intuition': 2069, 'simplify': 3706, 'explain': 1475, 'audience': 319, 'executives': 1441, 'canastota': 497, 'know': 2176, 'improved': 1922, 'autonomous': 338, 'sigint': 3690, 'isr': 2098, 'uas': 4177, 'cyber': 939, 'fly': 1589, 'original': 2745, 'rf': 3489, 'signals': 3693, 'procure': 3103, 'generate': 1683, 'curating': 922, 'throughput': 4077, 'hyper': 1870, 'tuning': 4167, 'strategies': 3874, 'thoroughly': 4073, 'characterizing': 573, 'integrate': 2018, 'chains': 552, 'pipelines': 2923, 'portable': 2978, 'raspberry': 3267, 'expect': 1451, 'means': 2413, 'incremental': 1941, 'government': 1719, 'investigation': 2074, 'classified': 614, 'citizenship': 605, 'nice': 2609, 'haves': 1792, 'npl': 2651, 'sequence': 3647, 'tidyverse': 4080, 'linux': 2293, 'radio': 3253, 'amateur': 157, 'agile': 118, 'think': 4069, 'northeast': 2636, 'discovery': 1141, 'persons': 2876, 'prohibits': 3138, 'discrimination': 1144, 'military': 2483, 'unrelated': 4222, 'marital': 2385, 'genetic': 1687, 'senioremployment': 3639, 'timea': 4083, 'jr': 2142, 'jersey': 2116, 'nj': 2613, 'chubb': 594, 'tap': 3995, 'geared': 1676, 'acumen': 60, 'joining': 2127, 'chosen': 593, 'mentorship': 2443, 'peers': 2853, 'hackathons': 1772, 'combination': 691, 'intellectually': 2025, 'curious': 925, 'curiosity': 924, 'resolve': 3435, 'decrease': 1006, 'adaptable': 64, 'overcome': 2762, 'starter': 3836, 'initiatives': 1977, 'estimation': 1401, 'recommend': 3313, 'gradient': 1728, 'topic': 4108, 'modelling': 2518, 'entity': 1356, 'rnn': 3497, 'cntk': 653, 'shiny': 3683, 'bokeh': 438, 'manipulation': 2371, 'energetic': 1319, 'multitude': 2562, 'possess': 2985, 'persuade': 2878, 'bonus': 439, 'parallel': 2795, 'pyspark': 3206, 'databricks': 962, 'eeo': 1257, 'laws': 2207, 'conditions': 774, 'religious': 3377, 'creed': 902, 'expression': 1488, 'ancestry': 187, 'pregnancy': 3035, 'characteristic': 571, 'assign': 282, 'compensate': 730, 'retain': 3463, 'unlawful': 4213, 'harassment': 1784, 'retaliation': 3465, 'locations': 2313, 'pmtechnical': 2964, 'disciplinebusiness': 1133, 'industryit': 1956, 'programnetworking': 3134, 'hackathonsyou': 1773, 'challengesyou': 561, 'etcunderstanding': 1404, 'scalaproficiency': 3560, 'languagesability': 2196, 'audienceexperience': 320, 'pythonc': 3210, 'experienceuse': 1464, 'codeenergetic': 658, 'requiredexperience': 3417, 'baltimore': 364, 'md': 2411, 'citizens': 604, 'combine': 693, 'ensemble': 1343, 'apache': 208, 'seaborn': 3597, 'mllib': 2507, 'notebooks': 2643, 'jupyter': 2153, 'zeppelin': 4411, 'notebook': 2642, 'microstrategy': 2475, 'looker': 2321, 'mind': 2488, 'huge': 1861, 'bsc': 469, 'ba': 352, 'fluently': 1587, 'facts': 1508, 'deadline': 991, 'closure': 645, 'persuasion': 2879, 'facilitation': 1505, 'thrive': 4076, 'rapidly': 3265, 'evolving': 1421, 'algorithmscombine': 138, 'modelingpresent': 2517, 'techniquespropose': 4022, 'challengescollaborate': 560, 'scienceexperience': 3579, 'miningexperience': 2495, 'mllibexperience': 2508, 'acumenproficiency': 61, 'analysisproblem': 176, 'aptitudeexperience': 249, 'skillsbsc': 3723, 'preferredexcellent': 3031, 'writingshould': 4394, 'descriptive': 1059, 'msc': 2553, 'fieldexperience': 1544, 'quantifying': 3232, 'crowd': 911, 'sourcing': 3769, 'protocolsstrong': 3164, 'toolssolid': 4107, 'perl': 2865, 'bash': 374, 'solid': 3750, 'sqlknowledge': 3812, 'techniquesexperience': 4021, 'choose': 592, 'upsexcellent': 4238, 'amazon': 158, 'minority': 2497, 'female': 1537, 'ai': 124, 'savvy': 3555, 'thinks': 4071, 'analytically': 181, 'believes': 396, 'measured': 2416, 'tested': 4050, 'reading': 3279, 'hard': 1785, 'mandate': 2366, 'innovate': 1978, 'unprecedented': 4220, 'rate': 3268, 'seemingly': 3617, 'endless': 1316, 'nlu': 2616, 'main': 2345, 'charge': 574, 'diving': 1169, 'root': 3515, 'testingpresent': 4052, 'backed': 355, 'coupled': 876, 'conclusions': 772, 'decisionscollaborate': 1003, 'effortshandle': 1270, 'competing': 734, 'customerscommunicate': 933, 'stack': 3821, 'redshift': 3334, 'demonstrable': 1030, 'dealing': 993, 'ambiguity': 159, 'prioritizing': 3074, 'delivering': 1024, 'dynamic': 1221, 'environmentfluency': 1367, 'foreign': 1604, 'languagemsc': 2194, 'detaildiving': 1073, 'visualizationsexperience': 4314, 'languag': 2192, 'livermore': 2301, 'llnl': 2305, 'safer': 3530, 'openings': 2705, 'dssi': 1216, 'really': 3286, 'allows': 150, 'acquired': 46, 'numerical': 2657, 'site': 3714, 'seminars': 3634, 'presented': 3054, 'grad': 1724, 'standing': 3833, 'certain': 545, 'suited': 3926, 'competition': 735, 'eagerness': 1224, 'frequentist': 1632, 'inverse': 2072, 'quantification': 3229, 'multimedia': 2558, 'signal': 3692, 'combinatorics': 692, 'graph': 1738, 'gpa': 1721, 'placement': 2935, 'exam': 1425, 'examination': 1426, 'pass': 2822, 'marijuana': 2384, 'contractor': 827, 'round': 3516, 'breaks': 457, 'unofficial': 4218, 'transcripts': 4127, 'east': 1231, 'premier': 3036, 'nuclear': 2653, 'nnsa': 2619, 'strengthening': 3883, 'respond': 3442, 'billion': 417, 'employing': 1299, 'analysisbayesian': 169, 'frameworksinverse': 1625, 'quantificationmachine': 3230, 'learningcomputer': 2230, 'visionmultimedia': 4304, 'processingcombinatorics': 3097, 'algorithmsgraph': 140, 'analysismodeling': 174, 'simulationmust': 3708, 'advent': 96, 'genomic': 1691, 'sequencing': 3649, 'finally': 1560, 'makeup': 2357, 'exist': 1445, 'tempus': 4038, 'greatest': 1747, 'promise': 3142, 'cancer': 498, 'diseases': 1149, 'lies': 2269, 'attributes': 317, 'redefine': 3331, 'imaging': 1900, 'precision': 3018, 'wrangle': 4389, 'sparse': 3773, 'interrogate': 2059, 'robustness': 3504, 'validity': 4263, 'sample': 3541, 'stability': 3818, 'bioinformatics': 424, 'supervised': 3940, 'unsupervised': 4225, 'mac': 2338, 'minimal': 2492, 'things': 4068, 'seamlessly': 3599, 'kaggle': 2158, 'competitions': 736, 'kernels': 2166, 'plot': 2954, 'ly': 2335, 'flask': 1578, 'dask': 952, 'iterations': 2106, 'medicinedesign': 2425, 'algorithmswrangle': 144, 'opportunitiesinterrogate': 2716, 'stabilitydocument': 3819, 'stakeholdersdegree': 3830, 'dataproficient': 970, 'sqlexperience': 3811, 'notebooksexperience': 2644, 'boostingexperience': 444, 'environmentoutstanding': 1371, 'skillsself': 3729, 'directiona': 1119, 'dothrive': 1195, 'seamlesslyexperience': 3600, 'audienceskaggle': 324, 'recordexperience': 3324, 'dataexperience': 965, 'architectureexperience': 254, 'daskexperience': 953, 'iter': 2103, 'swe': 3964, 'santa': 3546, 'clara': 608, 'valley': 4264, 'iphone': 2090, 'ipad': 2089, 'watch': 4340, 'macintosh': 2341, 'quantities': 3234, 'talk': 3993, 'substantial': 3914, 'grasp': 1745, 'especially': 1392, 'columnar': 688, 'vertica': 4291, 'hardware': 1787, 'phenomena': 2891, 'accumulated': 34, 'validate': 4260, 'hypotheses': 1871, 'devices': 1093, 'audit': 325, 'mclean': 2410, 'va': 4257, 'america': 163, 'virginia': 4300, 'capital': 507, 'founder': 1620, 'chairman': 555, 'chief': 587, 'officer': 2686, 'richard': 3490, 'fairbank': 1512, 'bringing': 462, 'ingenuity': 1971, 'simplicity': 3705, 'humanity': 1863, 'banking': 367, 'enjoy': 1337, 'exhibit': 1444, 'succeeding': 3917, 'guided': 1766, 'valued': 4268, 'powered': 3005, 'perspective': 2877, 'teamwork': 4015, 'respect': 3441, 'superior': 3938, 'elevate': 1280, 'obsess': 2671, 'thing': 4067, 'associates': 297, 'humility': 1866, 'realize': 3284, 'dreams': 1206, 'quest': 3239, 'tasked': 3999, 'transforming': 4135, 'enterprise': 1350, 'marry': 2393, 'result': 3452, 'intuitive': 2070, 'aren': 259, 'siloed': 3701, 'instead': 1999, 'invent': 2071, 'continuous': 824, 'priority': 3075, 'derive': 1053, 'detect': 1076, 'anomalies': 197, 'designed': 1062, 'improvement': 1923, 'eager': 1223, 'published': 3192, 'ask': 274, 'openly': 2706, 'disruptive': 1154, 'minded': 2489, 'linguistics': 2287, 'scraping': 3590, 'kibana': 2168, 'syntactic': 3971, 'parsing': 2804, 'semantic': 3628, 'fuzzy': 1662, 'textual': 4056, 'entailment': 1348, 'embedding': 1286, 'certificates': 547, 'sponsoring': 3803, 'authorization': 330, 'fieldat': 1543, 'linguisticsat': 2288, 'scrapingat': 3591, 'scalaat': 3558, 'azureph': 351, 'environmenthold': 1368, 'certif': 546, 'guild': 1769, 'advisor': 106, 'bellevue': 397, 'ne': 2578, 'liberal': 2265, 'arts': 271, 'enrolls': 1342, 'graduation': 1733, 'placed': 2934, 'trusted': 4159, 'encounter': 1310, 'streamlined': 3877, 'completing': 746, 'builds': 476, 'registrar': 3353, 'expectations': 1452, 'met': 2449, 'advisory': 107, 'pcas': 2845, 'realistic': 3282, 'finishing': 1571, 'introduces': 2065, 'dedicated': 1008, 'counselor': 871, 'keeps': 2161, 'apprised': 236, 'looks': 2323, 'efficiency': 1266, 'resulting': 3453, 'registration': 3354, 'moderator': 2528, 'facilitates': 1504, 'enters': 1351, 'receives': 3301, 'tracks': 4115, 'appropriately': 241, 'stressful': 3884, 'normal': 2633, 'profit': 3127, 'nebraska': 2581, 'south': 3770, 'omaha': 2692, 'vibrant': 4296, 'prosperous': 3158, 'metropolitan': 2465, 'nearly': 2580, 'residents': 3432, 'forbes': 1601, 'ranked': 3261, 'affordable': 112, 'live': 2300, 'cities': 602, 'interactive': 2033, 'tuition': 4166, 'home': 1838, 'jeanne': 2114, 'clery': 628, 'disclosure': 1136, 'crime': 904, 'alcohol': 132, 'abuse': 8, 'prevention': 3063, 'daapp': 945, 'prospective': 3157, 'clicking': 630, 'links': 2292, 'requesting': 3409, 'calling': 489, 'emailing': 1284, 'dso': 1215, 'environmentmay': 1370, 'denver': 1037, 'transamerica': 4125, 'formulate': 1612, 'cleanse': 622, 'refine': 3342, 'lucid': 2330, 'assumptions': 299, 'constructive': 803, 'criticism': 908, 'assessing': 279, 'customizations': 934, 'harden': 1786, 'quantiative': 3228, 'advances': 92, 'environmentdesign': 1366, 'opportunitieswork': 2717, 'modelinganalyze': 2516, 'developmentwrite': 1090, 'actcollaborate': 49, 'improvementbuild': 1924, 'performanceharden': 2860, 'timemasters': 4089, 'fieldthree': 1550, 'roleexcellent': 3508, 'algorithmsfluent': 139, 'rexperience': 3488, 'tensorflowsolid': 4044, 'designexcellent': 1064, 'ecosystemexcellent': 1242, 'storywilling': 3872, 'learningexperience': 2231, 'understandingin': 4199, 'learningoffice': 2233, 'environ': 1361, 'midlevel': 2479, 'midlevelemployment': 2480, 'jose': 2131, 'excited': 1437, 'appzen': 244, 'grow': 1755, 'productive': 3113, 'learningbuilding': 2229, 'modelsin': 2524, 'modelsunsupervised': 2525, 'pythonself': 3215, 'dayexcellent': 986, 'experiencephd': 1461, 'relatedexperience': 3367, 'los': 2324, 'altos': 155, 'datatron': 981, 'trivial': 4155, 'sci': 3577, 'kit': 2173, 'pto': 3186, 'stocked': 3864, 'snack': 3741, 'bar': 368, 'happy': 1783, 'outings': 2751, 'parking': 2803, 'onsite': 2700, 'clipper': 641, 'gym': 1770, 'reimbursement': 3361, 'discriminate': 1143, 'modelsdata': 2521, 'methodspredictive': 2462, 'datasetsexperience': 978, 'modelsexcellent': 2522, 'piggood': 2920, 'pytorchexperience': 3218, 'productionmedical': 3112, 'visionptoopportunities': 4305, 'growthlunch': 1760, 'dailyalways': 947, 'barfriday': 370, 'hoursmonthly': 1848, 'outingsfree': 2752, 'onsiteclipper': 2701, 'benefitsgym': 402, 'reimburs': 3360, 'angeles': 189, 'innovating': 1980, 'ucla': 4180, 'solved': 3756, 'linked': 2290, 'leverages': 2259, 'outputs': 2756, 'naïve': 2577, 'validation': 4262, 'ipython': 2091, 'shell': 3680, 'ehr': 1271, 'explaining': 1476, 'limitations': 2277, 'inferences': 1958, 'actual': 59, 'metadata': 2452, 'coaching': 654, 'consensus': 789, 'nesap': 2595, 'nersc': 2594, 'lbl': 2211, 'taleo': 3992, 'careersection': 516, 'jobdetail': 2122, 'ftl': 1642, 'supercomputing': 3936, 'scalable': 3559, 'acknowledged': 43, 'enabling': 1307, 'supercomputers': 3935, 'heterogeneous': 1811, 'cpu': 888, 'gpu': 1722, 'cray': 893, 'supercomputer': 3934, 'perlmutter': 2866, 'optimized': 2722, 'amd': 162, 'nvidia': 2661, 'speed': 3796, 'interconnect': 2034, 'flash': 1577, 'adapted': 65, 'efficiently': 1268, 'ones': 2697, 'focused': 1591, 'workload': 4384, 'today': 4096, 'exponentially': 1484, 'computationally': 761, 'observational': 2670, 'eos': 1377, 'synchrotron': 3969, 'telescopes': 4033, 'microscopes': 2472, 'particle': 2810, 'accelerators': 15, 'genome': 1689, 'sequencers': 3648, 'seamless': 3598, 'superfacility': 3937, 'composed': 757, 'pushes': 3203, 'vendor': 4280, 'ways': 4343, 'parallelization': 2797, 'bandwidth': 366, 'capitalize': 508, 'ecosystems': 1244, 'dl': 1173, 'refereed': 3337, 'vendors': 4281, 'enrichment': 1338, 'labs': 2188, 'demonstrably': 1031, 'productively': 3114, 'balancing': 363, 'hll': 1828, 'julia': 2148, 'corresponding': 865, 'parallelizing': 2798, 'parallelism': 2796, 'acceleration': 14, 'hlls': 1829, 'numba': 2654, 'dataframes': 969, 'jl': 2119, 'mpi': 2551, 'py': 3205, 'container': 816, 'docker': 1177, 'adapting': 66, 'introducing': 2066, 'synergistic': 3970, 'welcome': 4357, 'references': 3340, 'pursue': 3200, 'look': 2320, 'remote': 3380, 'let': 2249, 'precisiongx': 3019, 'insurers': 2016, 'institutes': 2001, 'governments': 1720, 'corporations': 860, 'insurance': 2011, 'claims': 607, 'try': 4163, 'untangle': 4226, 'engine': 1325, 'sqlalchemy': 3810, 'server': 3655, 'fluency': 1585, 'startup': 3839, 'capability': 503, 'windows': 4369, 'interesting': 2037, 'audiencesfluency': 323, 'guidancecapability': 1764, 'remotelinux': 3381, 'experienceprior': 1462, 'processinghealthtravel': 3100, 'simple': 3704, 'complicated': 751, 'practicality': 3011, 'beats': 385, 'purity': 3197, 'silently': 3698, 'explicitly': 1478, 'silenced': 3697, 'zen': 4410, 'humans': 1864, 'weren': 4361, 'understandable': 4197, 'integrating': 2020, 'apis': 211, 'easy': 1232, 'reliable': 3375, 'touch': 4111, 'parts': 2819, 'kinds': 2171, 'indexing': 1947, 'normalizing': 2634, 'crafting': 891, 'kafka': 2157, 'druid': 1213, 'postgresql': 2997, 'elastic': 1272, 'frontend': 1638, 'kubernetes': 2182, 'lgbtq': 2262, 'fundamentals': 1653, 'cs': 912, 'redis': 3332, 'gcp': 1675, 'authorized': 331, 'vacation': 4258, 'fsa': 1639, 'transit': 4136, 'sf': 3672, 'holiday': 1836, 'days': 987, 'proving': 3178, 'legally': 2245, 'usa': 4242, 'grmwa': 1749, 'iwlyindexing': 2108, 'analyticsoptimizing': 183, 'processingscaling': 3102, 'pipelinescrafting': 2924, 'applicationsleveraging': 226, 'toolsmaking': 4106, 'technologiespython': 4027, 'processinggo': 3099, 'apishigh': 212, 'databasesall': 960, 'kubernetesstrong': 2183, 'developmentfamiliar': 1088, 'redisuse': 3333, 'componentscloud': 756, 'platformauthorized': 2949, 'statesopen': 3846, 'policyhealth': 2968, 'insurancedental': 2012, 'insurancelife': 2014, 'insuranceshort': 2015, 'insurancehealth': 2013, 'fsatransit': 1641, 'fsafree': 1640, 'officeflexible': 2684, 'hoursholiday': 1847, 'joint': 2128, 'jgi': 2118, 'annotation': 191, 'metabolic': 2451, 'reconstruction': 3321, 'pathway': 2831, 'ontologies': 2702, 'pairwise': 2787, 'phylogenetic': 2906, 'biologists': 426, 'bioinformaticists': 423, 'microbial': 2467, 'genomes': 1690, 'metagenomes': 2453, 'reaction': 3274, 'phenotype': 2892, 'collating': 672, 'manual': 2375, 'curation': 923, 'microbiomes': 2468, 'img': 1901, 'knowledgebase': 2179, 'kbase': 2159, 'portability': 2977, 'newly': 2604, 'troubleshoot': 4156, 'annotations': 192, 'reconstructions': 3322, 'thousands': 4075, 'qualitative': 3225, 'sc': 3556, 'unix': 4212, 'documenting': 1184, 'bug': 473, 'creativity': 899, 'solo': 3751, 'protein': 3161, 'biochemical': 421, 'reactions': 3275, 'pathways': 2832, 'conversion': 840, 'classification': 613, 'depend': 1041, 'mitchell': 2504, 'walnut': 4333, 'creek': 903, 'mannerwork': 2374, 'metagenomesfacilitate': 2454, 'answeriq': 203, 'dedicate': 1007, 'ticket': 4079, 'classifying': 617, 'customized': 935, 'dataset': 974, 'graphical': 1739, 'mxnet': 2563, 'semantics': 3629, 'question': 3240, 'answering': 201, 'preferredstrong': 3034, 'answeringstrong': 202, 'busi': 479, 'mountain': 2549, 'earn': 1227, 'middle': 2478, 'junior': 2151, 'corporate': 859, 'profitable': 3128, 'expand': 1449, 'began': 388, 'grown': 1757, 'inception': 1928, 'heart': 1802, 'downtown': 1198, 'steps': 3860, 'caltrain': 491, 'station': 3848, 'interview': 2062, 'brightest': 460, 'silicon': 3699, 'profits': 3129, 'charitable': 576, 'poorest': 2972, 'regions': 3352, 'traffic': 4118, 'websites': 4349, 'slice': 3734, 'dice': 1099, 'account': 30, 'querying': 3238, 'digging': 1109, 'stories': 3870, 'telling': 4035, 'equally': 1380, 'tens': 4041, 'queue': 3244, 'bigquery': 416, 'etling': 1408, 'proactively': 3082, 'voice': 4324, 'owner': 2773, 'kpis': 2181, 'logistic': 2316, 'regressions': 3356, 'recurrent': 3330, 'nets': 2597, 'portion': 2980, 'pick': 2915, 'etls': 1409, 'juggling': 2147, 'qlikview': 3220, 'singular': 3712, 'decomposition': 1005, 'descent': 1055, 'groovy': 1751, 'compiled': 739, 'casual': 528, 'financially': 1563, 'stable': 3820, 'visitors': 4309, 'holidays': 1837, 'subsidized': 3913, 'kitchen': 2174, 'beverages': 409, 'sponsored': 3802, 'memberships': 2433, 'celebrate': 535, 'complies': 752, 'governing': 1718, 'additionally': 72, 'arrest': 264, 'uoejtog': 4227, 'ioas': 2086, 'plussolid': 2961, 'databasesexperience': 961, 'pipelinessolid': 2926, 'visualizationstrong': 4315, 'datacomfortable': 963, 'projectsbs': 3141, 'fieldsgood': 1547, 'passionatereporting': 2825, 'methodsknowledge': 2461, 'practicesknowledge': 3014, 'phdhelp': 2890, 'itenjoy': 2102, 'environmentjoin': 1369, 'companyhelp': 726, 'directionsee': 1121, 'yearshave': 4406, 'monthwork': 2545, 'memberscompetitive': 2432, 'annuallycomprehensive': 196, 'plansfully': 2941, 'passtuition': 2827, 'programsstocked': 3136, 'providedcompany': 3173, 'membershipscompany': 2434, 'qualificationsmaster': 3223, 'preferredproficiency': 3033, 'preferredexperience': 3032, 'preferreddemonstrated': 3030, 'requiredexcellent': 3416, 'executivesexperience': 1442, 'modelshigh': 2523, 'functioning': 1650, 'rapport': 3266, 'receiving': 3302, 'delegating': 1019, 'replacing': 3390, 'keeping': 2160, 'learnexperience': 2223, 'datasetsa': 976, 'chicago': 586, 'il': 1893, 'forefront': 1603, 'capaxglobal': 506, 'presence': 3050, 'industries': 1954, 'footprint': 1600, 'comes': 698, 'markets': 2390, 'exploring': 1483, 'connecting': 788, 'capax': 505, 'exposed': 1485, 'spectrum': 3793, 'visibility': 4302, 'fraud': 1628, 'procurement': 3104, 'supply': 3949, 'chain': 551, 'philadelphia': 2893, 'aligned': 145, 'ambiguous': 160, 'charters': 578, 'clearly': 626, 'risks': 3495, 'disciplinary': 1131, 'deliverables': 1022, 'glms': 1703, 'dplyr': 1199, 'caret': 519, 'versioning': 4290, 'eda': 1245, 'lay': 2208, 'sr': 3815, 'remotetravel': 3382, 'traverse': 4146, 'mi': 2466, 'hagerty': 1777, 'provider': 3174, 'classic': 612, 'car': 510, 'valuation': 4266, 'roadside': 3501, 'cars': 522, 'faced': 1502, 'enthusiasts': 1353, 'developers': 1085, 'creation': 897, 'productionize': 3111, 'output': 2755, 'proofs': 3149, 'concept': 766, 'globally': 1705, 'iterate': 2104, 'versed': 4289, 'presto': 3061, 'graphx': 1743, 'ex': 1423, 'xgboost': 4400, 'overfitting': 2763, 'authoring': 329, 'io': 2085, 'moa': 2511, 'mlpack': 2509, 'imperative': 1910, 'declarative': 1004, 'reasoning': 3294, 'figuring': 1553, 'aa': 0, 'wild': 4366, 'enthusiastic': 1352, 'achievers': 40, 'optimizationlead': 2720, 'graphxdeep': 1744, 'pythonbackground': 3209, 'pythonstrong': 3216, 'wall': 4332, 'journal': 2132, 'read': 3277, 'acknowledge': 42, 'dow': 1196, 'jones': 2129, 'cookie': 845, 'transfers': 4130, 'described': 1056, 'juniors': 2152, 'seniors': 3640, 'immersed': 1905, 'pulitzer': 3194, 'prize': 3080, 'winning': 4371, 'newsroom': 2607, 'journalists': 2135, 'interns': 2049, 'pitch': 2929, 'reported': 3392, 'graphics': 1740, 'edited': 1247, 'photo': 2900, 'wrote': 4396, 'colorful': 687, 'produced': 3106, 'videos': 4298, 'curated': 921, 'journalism': 2133, 'strive': 3886, 'welcoming': 4358, 'storied': 3869, 'immediately': 1903, 'investigations': 2075, 'reporters': 3393, 'billions': 418, 'rows': 3522, 'battling': 377, 'foia': 1595, 'officers': 2687, 'poorly': 2973, 'formed': 1610, 'ferrets': 1539, 'visualizing': 4318, 'scrape': 3589, 'mold': 2533, 'understands': 4200, 'dedication': 1009, 'puts': 3204, 'heavy': 1803, 'journalistic': 2134, 'exceptional': 1435, 'outlet': 2753, 'freelancer': 1631, 'pdfs': 2848, 'isnov': 2097, 'timing': 4092, 'internships': 2051, 'june': 2150, 'dates': 984, 'circumstances': 599, 'contacted': 813, 'rolling': 3513, 'selections': 3624, 'mid': 2477, 'wsj': 4397, 'li': 2263, 'ja': 2109, 'wsjintern': 4398, 'newsworthy': 2608, 'vets': 4295, 'reasonable': 3293, 'accommodation': 23, 'disabilities': 1125, 'talentresourceteam': 3991, 'dowjones': 1197, 'category': 530, 'commentary': 705, 'engages': 1323, 'mobile': 2512, 'heritage': 1810, 'politics': 2970, 'lifestyle': 2272, 'sports': 3804, 'holds': 1835, 'corp': 858, 'nasdaq': 2570, 'nws': 2663, 'nwsa': 2664, 'asx': 301, 'nwslv': 2665, 'workday': 4377, 'homepage': 1839, 'thank': 4058, 'id': 1879, 'broomfield': 467, 'cth': 916, 'scoring': 3588, 'elasticsearch': 1273, 'kellyjo': 2163, 'kiess': 2169, 'comcentric': 696, 'sponsor': 3801, 'citizen': 603, 'green': 1748, 'kelly': 2162, 'jo': 2120, 'techniquesdata': 4020, 'methodsextending': 2460, 'neededenhancing': 2586, 'systemsprocessing': 3977, 'analysisdoing': 172, 'mannercreating': 2373, 'performanceexcellent': 2859, 'elasticsearchgood': 1274, 'minted': 2498, 'specialist': 3776, 'fitting': 1575, 'ir': 2092, 'architect': 251, 'zero': 4412, 'foster': 1616, 'relatively': 3372, 'csv': 915, 'txt': 4173, 'xml': 4402, 'xlsx': 4401, 'ec': 1233, 'sklearn': 3730, 'dev': 1082, 'particularly': 2812, 'lucerne': 2329, 'switzerland': 3965, 'occasional': 2674, 'fine': 1566, 'brief': 458, 'starz': 3840, 'skilled': 3721, 'monica': 2537, 'entailing': 1347, 'clickstream': 631, 'size': 3717, 'brainstorm': 449, 'visualize': 4316, 'utilizing': 4255, 'sagemaker': 3532, 'ee': 1255, 'environmentstrong': 1374, 'dataexposure': 967, 'processingautomation': 3096, 'servicemachine': 3659, 'mlexperience': 2506, 'learningb': 2227, 'houston': 1851, 'tx': 4172, 'hadoopminimum': 1776, 'attitudeshare': 313, 'fort': 1613, 'lauderdale': 2203, 'fl': 1576, 'blazing': 432, 'trail': 4119, 'questioning': 3241, 'quo': 3249, 'algorithmic': 135, 'attribution': 318, 'automotive': 337, 'advertisers': 100, 'rated': 3269, 'genius': 1688, 'tank': 3994, 'wear': 4344, 'hats': 1791, 'fair': 1511, 'trusty': 4161, 'unlock': 4215, 'breaking': 456, 'gains': 1665, 'manufacturers': 2376, 'detecting': 1077, 'exploiting': 1479, 'feel': 1532, 'brainstorming': 450, 'whiteboard': 4362, 'silo': 3700, 'deriving': 1054, 'semi': 3632, 'manipulating': 2370, 'noisy': 2622, 'log': 2314, 'rapid': 3264, 'datastudio': 980, 'experienceperform': 1460, 'analysisgenerate': 173, 'environmentprepare': 1372, 'patternsdemonstrate': 2839, 'datasetsmust': 979, 'marketingdesigning': 2389, 'structuresexperience': 3892, 'statisticsexperience': 3853, 'pythondata': 3211, 'learningknowledge': 2232, 'profici': 3121, 'multioffice': 2560, 'charles': 577, 'river': 3496, 'firm': 1572, 'litigation': 2299, 'guides': 1768, 'businesses': 481, 'crai': 892, 'pharmaceutical': 2885, 'biotechnology': 430, 'diagnostics': 1096, 'device': 1092, 'optimum': 2725, 'patients': 2834, 'shareholders': 3677, 'associations': 298, 'earliest': 1225, 'therapy': 4064, 'fulfill': 1644, 'coe': 661, 'consultants': 805, 'pricing': 3065, 'personally': 2874, 'physician': 2910, 'promotional': 3146, 'possessing': 2986, 'contributors': 834, 'focusing': 1593, 'ascend': 272, 'streams': 3878, 'analyse': 165, 'mm': 2510, 'latentgold': 2201, 'spss': 3808, 'pharmetrics': 2887, 'truven': 4162, 'marketscan': 2391, 'xponent': 4403, 'visuals': 4320, 'collegial': 684, 'mathematically': 2402, 'proficiently': 3124, 'variate': 4270, 'vector': 4279, 'machines': 2340, 'qlik': 3219, 'symphony': 3967, 'optum': 2728, 'plantrak': 2943, 'ddd': 990, 'roledata': 3507, 'resultsdata': 3455, 'rolecandidates': 3506, 'streamscandidates': 3879, 'scientistimport': 3583, 'toolsbuild': 4104, 'pythonapply': 3208, 'clientswork': 637, 'environmentthis': 1375, 'rolemaster': 3510, 'fieldability': 1542, 'pythondemonstrate': 3212, 'audiencesability': 322, 'organizationability': 2736, 'problemsexperience': 3086, 'plusknowledge': 2960, 'pluscandidates': 2957, 'appli': 216, 'pasadena': 2821, 'holder': 1832, 'estimated': 1399, 'months': 2544, 'labor': 2186, 'iec': 1889, 'iot': 2088, 'nasa': 2569, 'darpa': 950, 'jpl': 2139, 'missions': 2503, 'holderoffer': 1833, 'screenthe': 3593, 'dallas': 948, 'century': 544, 'neiman': 2592, 'marcus': 2382, 'served': 3654, 'discerning': 1129, 'true': 4158, 'founders': 1621, 'omni': 2694, 'channel': 569, 'retailer': 3462, 'luxury': 2334, 'merchandise': 2444, 'distinctive': 1161, 'shopping': 3687, 'retail': 3461, 'bergdorf': 403, 'goodman': 1712, 'brand': 451, 'mytheresa': 2565, 'brands': 452, 'finest': 1568, 'apparel': 214, 'accessories': 21, 'jewelry': 2117, 'beauty': 387, 'décor': 1222, 'operates': 2709, 'affluent': 111, 'gateway': 1669, 'landmark': 2191, 'fifth': 1552, 'avenue': 343, 'price': 3064, 'cater': 531, 'upscale': 4237, 'ecommerce': 1236, 'neimanmarcus': 2593, 'bergdorfgoodman': 404, 'horchow': 1842, 'lastcall': 2200, 'cusp': 929, 'nmg': 2617, 'shop': 3686, 'anytime': 206, 'sell': 3627, 'investments': 2078, 'channels': 570, 'measurable': 2414, 'maintainable': 2348, 'retraining': 3469, 'ultimate': 4186, 'prototypes': 3166, 'select': 3620, 'parameters': 2800, 'retrain': 3468, 'extend': 1489, 'deployments': 1049, 'expanding': 1450, 'jenkins': 2115, 'gitflow': 1698, 'static': 3847, 'comments': 707, 'texas': 4054, 'irving': 2093, 'colwell': 689, 'blvd': 434, 'contributor': 833, 'pmdesigning': 2963, 'systemsrunning': 3979, 'experimentsimplementing': 1469, 'algorithmsstudy': 142, 'prototypesdesign': 3167, 'systemsresearch': 3978, 'toolsdevelop': 4105, 'requirementsselect': 3420, 'methodsrun': 2463, 'experimentsperform': 1470, 'resultstrain': 3459, 'necessaryextend': 2583, 'frameworkskeep': 1626, 'fieldproven': 1545, 'roleexperience': 3509, 'storesexpanding': 3868, 'applicationaws': 223, 'architecturedeep': 253, 'algorithmsability': 137, 'skillsoutstanding': 3728, 'plu': 2955, 'washington': 4339, 'dc': 989, 'unlocking': 4216, 'secrets': 3607, 'fascinated': 1518, 'increasingly': 1940, 'connected': 787, 'answers': 204, 'extracting': 1497, 'adversarial': 97, 'fairness': 1513, 'interpretability': 2054, 'whitepapers': 4363, 'interface': 2040, 'definition': 1015, 'digest': 1108, 'graphsexperience': 1742, 'algebra': 133, 'calculusability': 487, 'effectivelyability': 1262, 'taskspursuing': 4001, 'verificationexperience': 4285, 'methodsexperience': 2459, 'pipelinesexperience': 2925, 'tensorflowknowledge': 4043, 'empowers': 1303, 'fearlessly': 1523, 'urt': 4241, 'pittsburgh': 2932, 'symphonyrm': 3968, 'physicians': 2911, 'trained': 4121, 'suggestions': 3924, 'bakery': 361, 'square': 3814, 'pittsburg': 2931, 'resident': 3431, 'conceptualize': 768, 'efficacy': 1264, 'options': 2727, 'add': 67, 'svms': 3963, 'versatility': 4288, 'display': 1152, 'carefully': 518, 'succinctly': 3921, 'emr': 1304, 'dataconceptualize': 964, 'physicianswrite': 2912, 'algorithmstest': 143, 'efficacycommunicate': 1265, 'modelswork': 2526, 'quicklyexperience': 3247, 'datause': 982, 'resultsdetail': 3456, 'resultsolid': 3458, 'resultsexperience': 3457, 'predictionlist': 3023, 'rankunsupervised': 3263, 'emrexperience': 1305, 'samsung': 3544, 'galaxy': 1666, 'phones': 2899, 'appliances': 217, 'smart': 3737, 'refrigerators': 3345, 'tvs': 4171, 'washers': 4338, 'dryers': 1214, 'naturally': 2575, 'spoken': 3800, 'dialog': 1098, 'sra': 3816, 'reality': 3283, 'robotics': 3502, 'tier': 4081, 'siggraph': 3689, 'acm': 44, 'trans': 4124, 'cvpr': 938, 'iccv': 1877, 'iccp': 1876, 'spie': 3798, 'josa': 2130, 'lgbt': 2261, 'technologiesfacilitate': 4025, 'environmentcontribute': 1365, 'submissionsphd': 3908, 'fieldsr': 1548, 'technologiesr': 4028, 'fieldsstrong': 1549, 'technologieshands': 4026, 'toolkitsstrong': 4102, 'skillsgood': 3727, 'skil': 3719, 'authorship': 332, 'oakland': 2667, 'dayton': 988, 'oh': 2691, 'converting': 842, 'turnitin': 4170, 'extends': 1492, 'prevent': 3062, 'cheating': 580, 'predictions': 3024, 'style': 3902, 'consistency': 796, 'clarity': 609, 'vital': 4322, 'proprietary': 3155, 'labeled': 2185, 'instances': 1998, 'teammates': 4012, 'skillsets': 3724, 'autonomy': 339, 'owning': 2775, 'steer': 3858, 'owners': 2774, 'aggregation': 117, 'surfaced': 3954, 'blogs': 433, 'attending': 306, 'champion': 563, 'grows': 1758, 'rstudio': 3523, 'recommender': 3316, 'llc': 2304, 'palo': 2788, 'alto': 154, 'ayata': 349, 'strength': 3881, 'methodology': 2457, 'preprocess': 3045, 'dictionary': 1101, 'base': 372, 'property': 3152, 'casulity': 529, 'houstin': 1850, 'reimbursment': 3362, 'avaialble': 340, 'candidatelead': 500, 'operationsdevelop': 2714, 'applicationsbuild': 225, 'softwareinteract': 3746, 'clientsunderstand': 636, 'engineeringdocument': 1328, 'modelscommunicate': 2520, 'clientsproven': 635, 'datasetsdemonstrable': 977, 'recognitionexcellent': 3309, 'pythongood': 3214, 'processingphd': 3101, 'disciplinesat': 1135, 'intelligenceat': 2027, 'learningbackground': 2228, 'officeparallel': 2685, 'plusexpertise': 2959, 'gaps': 1668, 'inconsistencies': 1937, 'mindset': 2490, 'acting': 50, 'fosters': 1617, 'agility': 119, 'inspires': 1997, 'centricity': 543, 'timemaster': 4088, 'designexperience': 1065, 'ecosystemexperience': 1243, 'learningability': 2225, 'inconsistenciescommunication': 1938, 'insightsexcellent': 1993, 'skillsexcellent': 3725, 'learningattention': 2226, 'focusthe': 1594, 'environmentacting': 1363, 'collaborationaccountability': 668, 'solutionagility': 3753, 'quocustomer': 3250, 'customeroffice': 931}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V8FR7HGG_5aK",
        "colab_type": "code",
        "outputId": "1f67f7b7-7ebd-457f-ddf8-d0d5583ebd61",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "print(tfidf_vectorizer.idf_)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[4.23212105 3.72129543 1.52407085 ... 4.63758616 4.23212105 4.63758616]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LiHfFmUI_7u6",
        "colab_type": "code",
        "outputId": "b5ade155-4dc0-43b8-c944-e35ca26c901a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 426
        }
      },
      "source": [
        "# Convert Sparse Matrix to Pandas Dataframe to see the word frequencies.\n",
        "doc_term_matrix = tfidf_matrix.todense()\n",
        "df_temp_idf = pd.DataFrame(doc_term_matrix, \n",
        "                  columns=tfidf_vectorizer.get_feature_names()) \n",
        "df_temp_idf"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>aa</th>\n",
              "      <th>abilities</th>\n",
              "      <th>ability</th>\n",
              "      <th>able</th>\n",
              "      <th>abreast</th>\n",
              "      <th>absence</th>\n",
              "      <th>abstract</th>\n",
              "      <th>abundance</th>\n",
              "      <th>abuse</th>\n",
              "      <th>academia</th>\n",
              "      <th>academic</th>\n",
              "      <th>academics</th>\n",
              "      <th>academy</th>\n",
              "      <th>accelerate</th>\n",
              "      <th>acceleration</th>\n",
              "      <th>accelerators</th>\n",
              "      <th>acceptable</th>\n",
              "      <th>acceptance</th>\n",
              "      <th>accepted</th>\n",
              "      <th>access</th>\n",
              "      <th>accessible</th>\n",
              "      <th>accessories</th>\n",
              "      <th>accommodate</th>\n",
              "      <th>accommodation</th>\n",
              "      <th>accompany</th>\n",
              "      <th>accomplish</th>\n",
              "      <th>accomplishments</th>\n",
              "      <th>accordance</th>\n",
              "      <th>according</th>\n",
              "      <th>accordingly</th>\n",
              "      <th>account</th>\n",
              "      <th>accountability</th>\n",
              "      <th>accounts</th>\n",
              "      <th>accredited</th>\n",
              "      <th>accumulated</th>\n",
              "      <th>accuracy</th>\n",
              "      <th>accurate</th>\n",
              "      <th>accurately</th>\n",
              "      <th>achieve</th>\n",
              "      <th>achievement</th>\n",
              "      <th>...</th>\n",
              "      <th>word</th>\n",
              "      <th>work</th>\n",
              "      <th>workday</th>\n",
              "      <th>worked</th>\n",
              "      <th>workers</th>\n",
              "      <th>workflow</th>\n",
              "      <th>workflows</th>\n",
              "      <th>workforce</th>\n",
              "      <th>working</th>\n",
              "      <th>workload</th>\n",
              "      <th>workplace</th>\n",
              "      <th>works</th>\n",
              "      <th>world</th>\n",
              "      <th>worldwide</th>\n",
              "      <th>wrangle</th>\n",
              "      <th>write</th>\n",
              "      <th>writers</th>\n",
              "      <th>writes</th>\n",
              "      <th>writing</th>\n",
              "      <th>writingshould</th>\n",
              "      <th>written</th>\n",
              "      <th>wrote</th>\n",
              "      <th>wsj</th>\n",
              "      <th>wsjintern</th>\n",
              "      <th>www</th>\n",
              "      <th>xgboost</th>\n",
              "      <th>xlsx</th>\n",
              "      <th>xml</th>\n",
              "      <th>xponent</th>\n",
              "      <th>year</th>\n",
              "      <th>years</th>\n",
              "      <th>yearshave</th>\n",
              "      <th>yes</th>\n",
              "      <th>york</th>\n",
              "      <th>young</th>\n",
              "      <th>zen</th>\n",
              "      <th>zeppelin</th>\n",
              "      <th>zero</th>\n",
              "      <th>zr</th>\n",
              "      <th>zuckerberg</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.098459</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.184044</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.036770</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.067773</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.076722</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.034778</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.088884</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.033053</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.021214</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.044203</td>\n",
              "      <td>0.090518</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.027544</td>\n",
              "      <td>0.024171</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.025135</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.016902</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.022102</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.021160</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.025135</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.022102</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.065253</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.015986</td>\n",
              "      <td>0.087094</td>\n",
              "      <td>0.020429</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.034837</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.018611</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.067891</td>\n",
              "      <td>0.068259</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.022798</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.071059</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.046956</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.025726</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.034198</td>\n",
              "      <td>0.049119</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.058631</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.026094</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.036493</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.033632</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.04772</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.054270</td>\n",
              "      <td>0.05427</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.045382</td>\n",
              "      <td>0.045687</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.020127</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.034517</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.044108</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.028721</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.024990</td>\n",
              "      <td>0.064505</td>\n",
              "      <td>0.032338</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.076041</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.033366</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.023331</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.021502</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.032338</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.029209</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.022067</td>\n",
              "      <td>0.024044</td>\n",
              "      <td>0.042299</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.018362</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.013460</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>70</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.096006</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.036624</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.063695</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.06092</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.042749</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.028249</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.030953</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.026393</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.040310</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.029550</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>71</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.028582</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.062956</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.021593</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.055187</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.026343</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>72</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.058070</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.075093</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.019357</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.026295</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.054930</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.094402</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.024235</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>73</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.029993</td>\n",
              "      <td>0.077420</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.069646</td>\n",
              "      <td>0.058429</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.015444</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.028859</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.064622</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>74</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.111024</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.044043</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.037077</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.049001</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.030521</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.030521</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.034173</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>75 rows × 4415 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     aa  abilities   ability      able  ...  zeppelin  zero   zr  zuckerberg\n",
              "0   0.0   0.000000  0.098459  0.000000  ...       0.0   0.0  0.0         0.0\n",
              "1   0.0   0.044203  0.090518  0.000000  ...       0.0   0.0  0.0         0.0\n",
              "2   0.0   0.000000  0.022798  0.000000  ...       0.0   0.0  0.0         0.0\n",
              "3   0.0   0.000000  0.058631  0.000000  ...       0.0   0.0  0.0         0.0\n",
              "4   0.0   0.000000  0.024990  0.064505  ...       0.0   0.0  0.0         0.0\n",
              "..  ...        ...       ...       ...  ...       ...   ...  ...         ...\n",
              "70  0.0   0.000000  0.096006  0.000000  ...       0.0   0.0  0.0         0.0\n",
              "71  0.0   0.000000  0.000000  0.000000  ...       0.0   0.0  0.0         0.0\n",
              "72  0.0   0.000000  0.000000  0.058070  ...       0.0   0.0  0.0         0.0\n",
              "73  0.0   0.000000  0.029993  0.077420  ...       0.0   0.0  0.0         0.0\n",
              "74  0.0   0.000000  0.111024  0.000000  ...       0.0   0.0  0.0         0.0\n",
              "\n",
              "[75 rows x 4415 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xXocWluz83LM",
        "colab_type": "text"
      },
      "source": [
        "## 6.Dim Reduction\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SSn1LpY76NGd",
        "colab_type": "text"
      },
      "source": [
        "### SVD"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QhtO-jP_8boz",
        "colab_type": "text"
      },
      "source": [
        "Use dimensionality reduction with singular value decomposition (SVD, also known as latent semantic analysis, or LSA) on our TFIDF vectors, which can improve clustering performance."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hvp4RjoU8bN_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# setting the no of components for SVD\n",
        "i_components = 15\n",
        "\n",
        "# configuring SVD with i_components\n",
        "svd = TruncatedSVD(n_components=i_components)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2btOQHc7FJdV",
        "colab_type": "text"
      },
      "source": [
        "Project the tfidf vectors onto the first N principal components.\n",
        "Though this is significantly fewer features than the original tfidf vector,\n",
        "they are stronger features, and the accuracy is higher."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HwtHKeWbAsIk",
        "colab_type": "text"
      },
      "source": [
        "Since LSA/SVD results are not normalized, we have to do the normalization.\n",
        "    "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WbmbG08cAw8O",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "normalizer = Normalizer(copy=False)\n",
        "\n",
        "lsa = make_pipeline(svd, normalizer)\n",
        "\n",
        "lsa_result = lsa.fit_transform(tfidf_matrix)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wG5jJ5LMWhj0",
        "colab_type": "text"
      },
      "source": [
        "lsa.fit_transform\n",
        "*  Input: n_samples x n_features (75 x 4415)\n",
        "*  Output: n_samples x n_components (75 x 15)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aH1zGTli3g_x",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "03f72c22-0841-4caa-8883-ef2f85c663f8"
      },
      "source": [
        "print(tfidf_matrix.shape)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(75, 4415)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CncbfBq0W9Zu",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "2ecd089a-f92e-44e1-9fb9-2df0c3beb9cf"
      },
      "source": [
        "print(np.shape(lsa_result))"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(75, 15)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MC4UWkO7sZ7r",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_temp = pd.DataFrame(lsa_result)\n",
        "df_temp = df_temp.apply(lambda x: np.round(x, decimals=2))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8ay368-8s4z1",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197
        },
        "outputId": "f050ed6f-6e13-4502-dabf-d6de8985cbe0"
      },
      "source": [
        "df_temp.head()"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.57</td>\n",
              "      <td>0.50</td>\n",
              "      <td>-0.09</td>\n",
              "      <td>0.43</td>\n",
              "      <td>0.16</td>\n",
              "      <td>-0.04</td>\n",
              "      <td>-0.03</td>\n",
              "      <td>-0.13</td>\n",
              "      <td>-0.20</td>\n",
              "      <td>-0.29</td>\n",
              "      <td>-0.07</td>\n",
              "      <td>0.23</td>\n",
              "      <td>0.01</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>0.03</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.79</td>\n",
              "      <td>0.28</td>\n",
              "      <td>-0.15</td>\n",
              "      <td>0.29</td>\n",
              "      <td>-0.12</td>\n",
              "      <td>0.13</td>\n",
              "      <td>-0.03</td>\n",
              "      <td>-0.15</td>\n",
              "      <td>-0.24</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>-0.09</td>\n",
              "      <td>0.01</td>\n",
              "      <td>0.11</td>\n",
              "      <td>-0.12</td>\n",
              "      <td>-0.19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.64</td>\n",
              "      <td>0.04</td>\n",
              "      <td>-0.14</td>\n",
              "      <td>0.04</td>\n",
              "      <td>-0.49</td>\n",
              "      <td>0.27</td>\n",
              "      <td>-0.29</td>\n",
              "      <td>0.04</td>\n",
              "      <td>0.09</td>\n",
              "      <td>-0.01</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.05</td>\n",
              "      <td>0.11</td>\n",
              "      <td>0.27</td>\n",
              "      <td>-0.27</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.50</td>\n",
              "      <td>0.46</td>\n",
              "      <td>-0.19</td>\n",
              "      <td>0.48</td>\n",
              "      <td>0.10</td>\n",
              "      <td>-0.04</td>\n",
              "      <td>-0.00</td>\n",
              "      <td>-0.20</td>\n",
              "      <td>-0.22</td>\n",
              "      <td>-0.31</td>\n",
              "      <td>-0.12</td>\n",
              "      <td>0.23</td>\n",
              "      <td>0.05</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>-0.04</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.69</td>\n",
              "      <td>0.26</td>\n",
              "      <td>-0.10</td>\n",
              "      <td>0.36</td>\n",
              "      <td>0.31</td>\n",
              "      <td>0.06</td>\n",
              "      <td>0.06</td>\n",
              "      <td>-0.28</td>\n",
              "      <td>-0.15</td>\n",
              "      <td>-0.14</td>\n",
              "      <td>-0.11</td>\n",
              "      <td>0.19</td>\n",
              "      <td>-0.17</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>0.12</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     0     1     2     3     4     5   ...    9     10    11    12    13    14\n",
              "0  0.57  0.50 -0.09  0.43  0.16 -0.04  ... -0.29 -0.07  0.23  0.01 -0.02  0.03\n",
              "1  0.79  0.28 -0.15  0.29 -0.12  0.13  ... -0.02 -0.09  0.01  0.11 -0.12 -0.19\n",
              "2  0.64  0.04 -0.14  0.04 -0.49  0.27  ... -0.01  0.00  0.05  0.11  0.27 -0.27\n",
              "3  0.50  0.46 -0.19  0.48  0.10 -0.04  ... -0.31 -0.12  0.23  0.05 -0.02 -0.04\n",
              "4  0.69  0.26 -0.10  0.36  0.31  0.06  ... -0.14 -0.11  0.19 -0.17 -0.02  0.12\n",
              "\n",
              "[5 rows x 15 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e5h_SGNDx64K",
        "colab_type": "text"
      },
      "source": [
        "To avoide concat generating NAN we need to reset index, as these are not necessary"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m7Km7h8lxQbM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_temp.reset_index(drop=True, inplace=True)\n",
        "df_idf.reset_index(drop=True, inplace=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jys4dTNOtia1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_lsa = pd.concat([df_idf['text'],df_temp],axis=1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5WADmbuFtz1G",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197
        },
        "outputId": "502c66c9-569a-4e15-8c02-448595fb3cbe"
      },
      "source": [
        "df_lsa.head()"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>text</th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>institutional data and research analyst u berk...</td>\n",
              "      <td>0.57</td>\n",
              "      <td>0.50</td>\n",
              "      <td>-0.09</td>\n",
              "      <td>0.43</td>\n",
              "      <td>0.16</td>\n",
              "      <td>-0.04</td>\n",
              "      <td>-0.03</td>\n",
              "      <td>-0.13</td>\n",
              "      <td>-0.20</td>\n",
              "      <td>-0.29</td>\n",
              "      <td>-0.07</td>\n",
              "      <td>0.23</td>\n",
              "      <td>0.01</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>0.03</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>data science health innovation fellow job bids...</td>\n",
              "      <td>0.79</td>\n",
              "      <td>0.28</td>\n",
              "      <td>-0.15</td>\n",
              "      <td>0.29</td>\n",
              "      <td>-0.12</td>\n",
              "      <td>0.13</td>\n",
              "      <td>-0.03</td>\n",
              "      <td>-0.15</td>\n",
              "      <td>-0.24</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>-0.09</td>\n",
              "      <td>0.01</td>\n",
              "      <td>0.11</td>\n",
              "      <td>-0.12</td>\n",
              "      <td>-0.19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>machine learning postdoctoral fellow san franc...</td>\n",
              "      <td>0.64</td>\n",
              "      <td>0.04</td>\n",
              "      <td>-0.14</td>\n",
              "      <td>0.04</td>\n",
              "      <td>-0.49</td>\n",
              "      <td>0.27</td>\n",
              "      <td>-0.29</td>\n",
              "      <td>0.04</td>\n",
              "      <td>0.09</td>\n",
              "      <td>-0.01</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.05</td>\n",
              "      <td>0.11</td>\n",
              "      <td>0.27</td>\n",
              "      <td>-0.27</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>data analyst u berkeley ca data analyst u abou...</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.46</td>\n",
              "      <td>-0.19</td>\n",
              "      <td>0.48</td>\n",
              "      <td>0.10</td>\n",
              "      <td>-0.04</td>\n",
              "      <td>-0.00</td>\n",
              "      <td>-0.20</td>\n",
              "      <td>-0.22</td>\n",
              "      <td>-0.31</td>\n",
              "      <td>-0.12</td>\n",
              "      <td>0.23</td>\n",
              "      <td>0.05</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>-0.04</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>senior data systems analyst u berkeley ca seni...</td>\n",
              "      <td>0.69</td>\n",
              "      <td>0.26</td>\n",
              "      <td>-0.10</td>\n",
              "      <td>0.36</td>\n",
              "      <td>0.31</td>\n",
              "      <td>0.06</td>\n",
              "      <td>0.06</td>\n",
              "      <td>-0.28</td>\n",
              "      <td>-0.15</td>\n",
              "      <td>-0.14</td>\n",
              "      <td>-0.11</td>\n",
              "      <td>0.19</td>\n",
              "      <td>-0.17</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>0.12</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                text     0  ...    13    14\n",
              "0  institutional data and research analyst u berk...  0.57  ... -0.02  0.03\n",
              "1  data science health innovation fellow job bids...  0.79  ... -0.12 -0.19\n",
              "2  machine learning postdoctoral fellow san franc...  0.64  ...  0.27 -0.27\n",
              "3  data analyst u berkeley ca data analyst u abou...  0.50  ... -0.02 -0.04\n",
              "4  senior data systems analyst u berkeley ca seni...  0.69  ... -0.02  0.12\n",
              "\n",
              "[5 rows x 16 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e3AH2ZrsBk5e",
        "colab_type": "code",
        "outputId": "9b2e81d7-c6bd-41f9-ede8-c6b1b8f2a8a6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "explained_variance = svd.explained_variance_ratio_.sum() \n",
        "print(\"Explained variance of the SVD step: {}%\".format(int(explained_variance * 100)))\n"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Explained variance of the SVD step: 44%\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "83PkRYr-KV5t",
        "colab_type": "code",
        "outputId": "0dbafee0-99bc-40ab-f35e-539c19553580",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "len(svd.components_)"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "15"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EG2yMq9YQYy-",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 70
        },
        "outputId": "f3aa700d-514d-472b-ad07-6ce047a15799"
      },
      "source": [
        " print(sort(svd.explained_variance_ratio_)[::-1]) # sorted in descending order"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[0.0705984  0.05069666 0.04073402 0.03446569 0.03097469 0.02862975\n",
            " 0.02697261 0.02515586 0.02332906 0.02215172 0.02093624 0.01903978\n",
            " 0.01773784 0.01689105 0.01202285]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gnw9CPfEMhqO",
        "colab_type": "text"
      },
      "source": [
        "Use k-means to cluster the dimensionally-reduced data. Decide on the optimal number of clusters."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "82LB4UYeF2WM",
        "colab_type": "text"
      },
      "source": [
        "## 7.Classification "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R2hfaOagMpmZ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "i_clusters = 5\n",
        "\n",
        "# init: Method for initialization, defaults to ‘k-means++’:\n",
        "# ‘k-means++’ : selects initial cluster centers for k-mean clustering in a smart way to speed up convergence. \n",
        "\n",
        "# n_init : Number of time the k-means algorithm will be run with different centroid seeds. \n",
        "# The final results will be the best output of n_init consecutive runs in terms of inertia\n",
        "\n",
        "# max_iterint, default=300\n",
        "#Maximum number of iterations of the k-means algorithm for a single run.\n",
        "\n",
        "km = KMeans(n_clusters=i_clusters, init='k-means++', max_iter=100, n_init=1)\n",
        "\n",
        "kmeans = km.fit(lsa_result)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "txQhM4Wq_xpP",
        "colab_type": "text"
      },
      "source": [
        "### Exploring KMeans"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "k8nqe9qQT9rN",
        "colab_type": "text"
      },
      "source": [
        "Coordinates of cluster centers --> \n",
        "cluster_centers_ ndarray of shape (n_clusters, n_features). \n",
        "\n",
        "In our case: \n",
        "* n_clusters = 5 = i_clusters\n",
        "* n_features = 15 = dim of SVD/LSA\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_1CogAe4-6oZ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "4b4e7b94-6e40-4490-830c-36715bdfe91d"
      },
      "source": [
        "np.shape(kmeans.cluster_centers_)"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(5, 15)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JJ-nAiShULEZ",
        "colab_type": "code",
        "outputId": "5c536704-5b99-4804-e81b-80b79299d8bc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 461
        }
      },
      "source": [
        "kmeans.cluster_centers_"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[ 5.69154515e-01, -2.58813269e-01, -1.88271065e-02,\n",
              "        -8.88278833e-02, -6.32777547e-02, -1.16400325e-01,\n",
              "         5.85392462e-02, -9.04671237e-02,  7.10874771e-02,\n",
              "         2.26035893e-01,  3.73166972e-01,  5.68203966e-01,\n",
              "        -1.14762656e-01, -1.98130166e-01, -1.54996113e-02],\n",
              "       [ 6.55167251e-01, -1.57270614e-01, -9.70101821e-03,\n",
              "         2.94597526e-02,  1.43402463e-01,  1.19770343e-01,\n",
              "         2.01903472e-02, -8.43878879e-02,  6.85491641e-03,\n",
              "         5.63896476e-02, -4.16850519e-02, -1.05738333e-01,\n",
              "        -5.77536705e-02, -6.51820618e-02, -6.45240763e-03],\n",
              "       [ 4.19094418e-01,  4.93409822e-01,  5.91221173e-02,\n",
              "         6.81648230e-04,  4.26450615e-02, -6.64430792e-03,\n",
              "         6.31901184e-03, -5.26417727e-02, -6.67175598e-02,\n",
              "        -9.25901554e-02, -4.31010281e-02,  8.03329396e-02,\n",
              "         1.44594944e-02,  3.66568966e-04, -8.11349833e-03],\n",
              "       [ 6.63409102e-01, -1.75856729e-01, -2.32172947e-02,\n",
              "        -5.67642215e-02, -1.74212614e-01, -7.78840711e-02,\n",
              "         3.84117234e-03,  9.28961856e-02, -4.46449499e-02,\n",
              "        -2.96760192e-02,  5.44824585e-03, -3.50789420e-02,\n",
              "         4.46701752e-02,  1.76679552e-01, -5.02074665e-03],\n",
              "       [ 2.66336596e-01,  3.16232045e-01,  3.44803031e-02,\n",
              "         3.71815682e-01,  2.62187656e-02, -2.41996015e-01,\n",
              "         1.49097259e-02,  1.07093876e-01,  1.95906421e-01,\n",
              "         1.08082124e-01,  3.00612067e-01, -2.39884569e-01,\n",
              "        -8.83021637e-03,  3.78238767e-02,  5.63314828e-02]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XRdk20Pe_NaG",
        "colab_type": "text"
      },
      "source": [
        "We get labels for each row of the input data i.e. 75"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KS4plL7HUVMJ",
        "colab_type": "text"
      },
      "source": [
        "labels_ ndarray of shape (n_samples,)\n",
        "Labels of each point\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fn5yew1IGFhc",
        "colab_type": "code",
        "outputId": "92d250c3-eb74-42ab-b34a-10e133177160",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "kmeans.labels_.shape"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(75,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j9zFZX0vTlik",
        "colab_type": "code",
        "outputId": "82fb21c6-cb2d-4fbc-bf71-8b974fe8d9ba",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        }
      },
      "source": [
        "kmeans.labels_"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([2, 2, 3, 2, 2, 2, 2, 3, 2, 2, 4, 4, 2, 2, 2, 4, 2, 2, 2, 2, 2, 3,\n",
              "       4, 2, 3, 4, 1, 1, 1, 3, 3, 3, 2, 2, 1, 1, 1, 3, 1, 1, 3, 1, 1, 3,\n",
              "       4, 0, 1, 3, 3, 1, 1, 1, 3, 1, 3, 1, 1, 1, 1, 3, 3, 1, 3, 1, 2, 1,\n",
              "       1, 3, 3, 3, 3, 3, 1, 3, 0], dtype=int32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dauAG1ypU5sC",
        "colab_type": "code",
        "outputId": "1e3977d9-1b91-4e0c-e0db-8bc1ae06a04a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "set(kmeans.labels_)"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{0, 1, 2, 3, 4}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N0UEHX5pPvO0",
        "colab_type": "text"
      },
      "source": [
        "For better observation, let us make this array part of the data frame so that we can look at the cluster each row belongs to, in the same data frame:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iX5w6moujnll",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_lsa['Cluster'] = kmeans.labels_"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ld6DGkoUz2cS",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197
        },
        "outputId": "e6778872-5d43-4d2f-e35c-2c4248660f43"
      },
      "source": [
        "df_lsa.head()"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>text</th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "      <th>Cluster</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>institutional data and research analyst u berk...</td>\n",
              "      <td>0.57</td>\n",
              "      <td>0.50</td>\n",
              "      <td>-0.09</td>\n",
              "      <td>0.43</td>\n",
              "      <td>0.16</td>\n",
              "      <td>-0.04</td>\n",
              "      <td>-0.03</td>\n",
              "      <td>-0.13</td>\n",
              "      <td>-0.20</td>\n",
              "      <td>-0.29</td>\n",
              "      <td>-0.07</td>\n",
              "      <td>0.23</td>\n",
              "      <td>0.01</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>0.03</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>data science health innovation fellow job bids...</td>\n",
              "      <td>0.79</td>\n",
              "      <td>0.28</td>\n",
              "      <td>-0.15</td>\n",
              "      <td>0.29</td>\n",
              "      <td>-0.12</td>\n",
              "      <td>0.13</td>\n",
              "      <td>-0.03</td>\n",
              "      <td>-0.15</td>\n",
              "      <td>-0.24</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>-0.09</td>\n",
              "      <td>0.01</td>\n",
              "      <td>0.11</td>\n",
              "      <td>-0.12</td>\n",
              "      <td>-0.19</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>machine learning postdoctoral fellow san franc...</td>\n",
              "      <td>0.64</td>\n",
              "      <td>0.04</td>\n",
              "      <td>-0.14</td>\n",
              "      <td>0.04</td>\n",
              "      <td>-0.49</td>\n",
              "      <td>0.27</td>\n",
              "      <td>-0.29</td>\n",
              "      <td>0.04</td>\n",
              "      <td>0.09</td>\n",
              "      <td>-0.01</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.05</td>\n",
              "      <td>0.11</td>\n",
              "      <td>0.27</td>\n",
              "      <td>-0.27</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>data analyst u berkeley ca data analyst u abou...</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.46</td>\n",
              "      <td>-0.19</td>\n",
              "      <td>0.48</td>\n",
              "      <td>0.10</td>\n",
              "      <td>-0.04</td>\n",
              "      <td>-0.00</td>\n",
              "      <td>-0.20</td>\n",
              "      <td>-0.22</td>\n",
              "      <td>-0.31</td>\n",
              "      <td>-0.12</td>\n",
              "      <td>0.23</td>\n",
              "      <td>0.05</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>-0.04</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>senior data systems analyst u berkeley ca seni...</td>\n",
              "      <td>0.69</td>\n",
              "      <td>0.26</td>\n",
              "      <td>-0.10</td>\n",
              "      <td>0.36</td>\n",
              "      <td>0.31</td>\n",
              "      <td>0.06</td>\n",
              "      <td>0.06</td>\n",
              "      <td>-0.28</td>\n",
              "      <td>-0.15</td>\n",
              "      <td>-0.14</td>\n",
              "      <td>-0.11</td>\n",
              "      <td>0.19</td>\n",
              "      <td>-0.17</td>\n",
              "      <td>-0.02</td>\n",
              "      <td>0.12</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                text     0  ...    14  Cluster\n",
              "0  institutional data and research analyst u berk...  0.57  ...  0.03        2\n",
              "1  data science health innovation fellow job bids...  0.79  ... -0.19        2\n",
              "2  machine learning postdoctoral fellow san franc...  0.64  ... -0.27        3\n",
              "3  data analyst u berkeley ca data analyst u abou...  0.50  ... -0.04        2\n",
              "4  senior data systems analyst u berkeley ca seni...  0.69  ...  0.12        2\n",
              "\n",
              "[5 rows x 17 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TaXOSPDxUaqk",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "n_iter_ int\n",
        "Number of iterations run."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V08X_YKVUfdr",
        "colab_type": "code",
        "outputId": "ca737ff6-c4e3-4bc5-92ef-a3ea2b73de65",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "kmeans.n_iter_"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "4"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XEDRmsWnSxwW",
        "colab_type": "text"
      },
      "source": [
        "the distribution of cluster assignments"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JfVYUYeYAe9H",
        "colab_type": "code",
        "outputId": "df4531cd-e63b-44ef-e591-33369622a086",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 267
        }
      },
      "source": [
        "plt.hist(kmeans.labels_, bins=i_clusters)\n",
        "plt.show()"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAD7CAYAAABzGc+QAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/wOZyWvTghRBBwAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OWnBPOSm0Ir-",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "f9bac608-66c4-4d52-d8f6-9369fcdc8642"
      },
      "source": [
        "\n",
        "c = Counter(kmeans.labels_) \n",
        "print(c.items())"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "dict_items([(2, 20), (3, 23), (4, 6), (1, 24), (0, 2)])\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1WsJxcJAo3Fd",
        "colab_type": "text"
      },
      "source": [
        "## 8.Text Clusters"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-Ey54qhNI6Wp",
        "colab_type": "text"
      },
      "source": [
        "### Setting Up"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "G1lgzStfDsLD",
        "colab_type": "text"
      },
      "source": [
        "How do we visualize the terms - text has been converted to numbers during SVD/Dim Reduction "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tDvyxE7YFmIU",
        "colab_type": "text"
      },
      "source": [
        "For each cluster, get the text \n",
        "For each text, get the terms\n",
        "For each term, get the TF-IDF score\n",
        "where TF-IDF score will determine the FontSize"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7X7qP2EaKNb0",
        "colab_type": "text"
      },
      "source": [
        "Subsetting to two columns\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LByx9XsSJyQw",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_plot = df_lsa[['text','Cluster']]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nk6B3soAI2C8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197
        },
        "outputId": "8b7d4768-6c5a-4abc-a283-341b406e7aed"
      },
      "source": [
        "df_plot.head()"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>text</th>\n",
              "      <th>Cluster</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>institutional data and research analyst u berk...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>data science health innovation fellow job bids...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>machine learning postdoctoral fellow san franc...</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>data analyst u berkeley ca data analyst u abou...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>senior data systems analyst u berkeley ca seni...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                text  Cluster\n",
              "0  institutional data and research analyst u berk...        2\n",
              "1  data science health innovation fellow job bids...        2\n",
              "2  machine learning postdoctoral fellow san franc...        3\n",
              "3  data analyst u berkeley ca data analyst u abou...        2\n",
              "4  senior data systems analyst u berkeley ca seni...        2"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zZwHT7LZZ68T",
        "colab_type": "text"
      },
      "source": [
        "### Driver Code"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PvP4V9q468U4",
        "colab_type": "text"
      },
      "source": [
        "Code to test the function:\n",
        "cluster_to_image()\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zsmeAhrOz91a",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 107
        },
        "outputId": "40b37a81-602d-45a9-8cb5-4f320e0dcc28"
      },
      "source": [
        "list(df_plot.groupby('Cluster'))[0][1]"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>text</th>\n",
              "      <th>Cluster</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>45</th>\n",
              "      <td>data scientist bioinformatics denver co job de...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>74</th>\n",
              "      <td>data scientist denver co job description summa...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                 text  Cluster\n",
              "45  data scientist bioinformatics denver co job de...        0\n",
              "74  data scientist denver co job description summa...        0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BIERzOvCc5lJ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "383f0a41-5f26-4ebf-8b42-38a855c944ba"
      },
      "source": [
        "# cluster 0\n",
        "df_0 = list(df_plot.groupby('Cluster'))[0][1]\n",
        "\n",
        "# cluster 2\n",
        "df_2 = list(df_plot.groupby('Cluster'))[2][1]\n",
        "\n",
        "cluster_to_image(df_0,10)"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "#samples: 1, #features: 217\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<wordcloud.wordcloud.WordCloud at 0x7fd8d68ae668>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_xCgUaHQq1X_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 236
        },
        "outputId": "aa198c43-f8b0-4667-bd09-45cda06539c3"
      },
      "source": [
        "plt.imshow(cluster_to_image(df_0,10))\n",
        "plt.show()"
      ],
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "#samples: 1, #features: 217\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAADKCAYAAABe4wDhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9q5QOTprtcV7",
        "colab_type": "text"
      },
      "source": [
        "### Visualizing all clusters using subplots"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ojyz03xg9XYi",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "e18d11fd-63a4-4a92-fb0c-65f323c381ac"
      },
      "source": [
        "int(5/2)"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5sAjM0Ubur2F",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "cluster_groups = list(df_plot.groupby('Cluster'))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HUSwA3vD-Pm-",
        "colab_type": "text"
      },
      "source": [
        "Caveat for below - will miss the last odd numbered plot (for e.g. 5th Cluster is not plotted because of i_rows, i_cols arrangement)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f694SZlvtcAu",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 804
        },
        "outputId": "f76e7840-2472-47b0-911a-e2de07aa167c"
      },
      "source": [
        "#i_rows = 5\n",
        "#i_cols = 4\n",
        "\n",
        "i_rows = int(i_clusters/2)\n",
        "i_cols = 2\n",
        "\n",
        "\n",
        "figure, axes = plt.subplots(i_rows, i_cols, figsize=(20, 15))\n",
        "\n",
        "for r in range(i_rows):\n",
        "  for c in range(i_cols):\n",
        "    _, df_cluster = cluster_groups.pop(0)\n",
        "    wordcloud_image = cluster_to_image(df_cluster, max_words=10)\n",
        "    ax = axes[r][c]\n",
        "    ax.imshow(wordcloud_image,interpolation=\"bilinear\")\n",
        "    #ax.set_title(get_title(df_cluster), fontsize=20) ②\n",
        "    ax.set_xticks([])\n",
        "    ax.set_yticks([])\n",
        "    \n",
        "    \n",
        "plt.show()\n"
      ],
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "#samples: 1, #features: 217\n",
            "#samples: 1, #features: 2088\n",
            "#samples: 1, #features: 1993\n",
            "#samples: 1, #features: 2096\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABG0AAALMCAYAAABe/k+2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 1440x1080 with 4 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7uOthtI5Cqs1",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "\n",
        "## 8.Store for Later\n",
        "\n",
        "Store the output for later\n",
        "\n",
        "\n",
        "---\n",
        "\n"
      ]
    }
  ]
}