psychemedia/simpleWordcloud.ipynb

## simpleWordcloud.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['th@t',\n",
       " 'this',\n",
       " 'goodbye',\n",
       " 'cogdog',\n",
       " 'th@t',\n",
       " 'whenever',\n",
       " 'goodbye',\n",
       " 'whenever',\n",
       " 'th@t',\n",
       " 'wherever']"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Random word list generator\n",
    "\n",
    "from random import choices\n",
    "_words = [\"hello\", \"goodbye\", \"this\", \"th@t\", 'whenever', 'wherever', 'cogdog']\n",
    "choices(_words, k=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "try:\n",
    "    import pandas as pd\n",
    "except:\n",
    "    !pip install pandas "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>col1</th>\n",
       "      <th>col2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>th@t</td>\n",
       "      <td>hello</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>whenever</td>\n",
       "      <td>th@t</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>th@t</td>\n",
       "      <td>goodbye</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cogdog</td>\n",
       "      <td>whenever</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>wherever</td>\n",
       "      <td>th@t</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       col1      col2\n",
       "0      th@t     hello\n",
       "1  whenever      th@t\n",
       "2      th@t   goodbye\n",
       "3    cogdog  whenever\n",
       "4  wherever      th@t"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Create a simple dataframe with two random word lists\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "df = pd.DataFrame({'col1':choices(_words, k=100), 'col2':choices(_words+['gotcha'], k=100)})\n",
    "\n",
    "#We can save the data to a csv file...\n",
    "df.to_csv('mywords.csv', index=False)\n",
    "\n",
    "#Or preview it\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "col1,col2\r\n",
      "th@t,hello\r\n",
      "whenever,th@t\r\n"
     ]
    }
   ],
   "source": [
    "#Here's what it looks like as csv\n",
    "!head -n 3 mywords.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>col1</th>\n",
       "      <th>col2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>th@t</td>\n",
       "      <td>hello</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>whenever</td>\n",
       "      <td>th@t</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>th@t</td>\n",
       "      <td>goodbye</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cogdog</td>\n",
       "      <td>whenever</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>wherever</td>\n",
       "      <td>th@t</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       col1      col2\n",
       "0      th@t     hello\n",
       "1  whenever      th@t\n",
       "2      th@t   goodbye\n",
       "3    cogdog  whenever\n",
       "4  wherever      th@t"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#load the csv into another dataframe - to show we can\n",
    "df2 = pd.read_csv('mywords.csv')\n",
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "#Install wordcloud package\n",
    "try:\n",
    "    import wordcloud\n",
    "except:\n",
    "    !pip install wordcloud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "try:\n",
    "    import matplotlib\n",
    "except:\n",
    "    !pip install matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Required graphics package\n",
    "import matplotlib.pyplot as plt\n",
    "#...and magic to diplay results inline in the notebook...\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from wordcloud import WordCloud\n",
    "\n",
    "# Generate a word cloud image\n",
    "wordcloud = WordCloud().generate(' '.join(df2['col1'].tolist()))\n",
    "\n",
    "\n",
    "\n",
    "plt.imshow(wordcloud, interpolation='bilinear');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "wordcloud = WordCloud().generate(' '.join(df2['col2'].tolist()))\n",
    "\n",
    "plt.imshow(wordcloud, interpolation='bilinear');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you have sentences, they can be split...\n",
    "\n",
    "It's particularly easy if the split is regular. For example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>col1</th>\n",
       "      <th>col2</th>\n",
       "      <th>col3a</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>th@t</td>\n",
       "      <td>hello</td>\n",
       "      <td>@th@t/hello</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>whenever</td>\n",
       "      <td>th@t</td>\n",
       "      <td>@whenever/th@t</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>th@t</td>\n",
       "      <td>goodbye</td>\n",
       "      <td>@th@t/goodbye</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cogdog</td>\n",
       "      <td>whenever</td>\n",
       "      <td>@cogdog/whenever</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>wherever</td>\n",
       "      <td>th@t</td>\n",
       "      <td>@wherever/th@t</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       col1      col2             col3a\n",
       "0      th@t     hello       @th@t/hello\n",
       "1  whenever      th@t    @whenever/th@t\n",
       "2      th@t   goodbye     @th@t/goodbye\n",
       "3    cogdog  whenever  @cogdog/whenever\n",
       "4  wherever      th@t    @wherever/th@t"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['col3a'] = '@'+df['col1']+ '/' + df['col2']\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>col1</th>\n",
       "      <th>col2</th>\n",
       "      <th>col3a</th>\n",
       "      <th>col3b</th>\n",
       "      <th>col3c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>th@t</td>\n",
       "      <td>hello</td>\n",
       "      <td>@th@t/hello</td>\n",
       "      <td>@th@t</td>\n",
       "      <td>hello</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>whenever</td>\n",
       "      <td>th@t</td>\n",
       "      <td>@whenever/th@t</td>\n",
       "      <td>@whenever</td>\n",
       "      <td>th@t</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>th@t</td>\n",
       "      <td>goodbye</td>\n",
       "      <td>@th@t/goodbye</td>\n",
       "      <td>@th@t</td>\n",
       "      <td>goodbye</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cogdog</td>\n",
       "      <td>whenever</td>\n",
       "      <td>@cogdog/whenever</td>\n",
       "      <td>@cogdog</td>\n",
       "      <td>whenever</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>wherever</td>\n",
       "      <td>th@t</td>\n",
       "      <td>@wherever/th@t</td>\n",
       "      <td>@wherever</td>\n",
       "      <td>th@t</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       col1      col2             col3a      col3b     col3c\n",
       "0      th@t     hello       @th@t/hello      @th@t     hello\n",
       "1  whenever      th@t    @whenever/th@t  @whenever      th@t\n",
       "2      th@t   goodbye     @th@t/goodbye      @th@t   goodbye\n",
       "3    cogdog  whenever  @cogdog/whenever    @cogdog  whenever\n",
       "4  wherever      th@t    @wherever/th@t  @wherever      th@t"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We can split a string in a column and then expand it over a couple of columns\n",
    "df[['col3b','col3c']] = df['col3a'].str.split('/', 1, expand=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Want username without the `@`?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>col1</th>\n",
       "      <th>col2</th>\n",
       "      <th>col3a</th>\n",
       "      <th>col3b</th>\n",
       "      <th>col3c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>th@t</td>\n",
       "      <td>hello</td>\n",
       "      <td>@th@t/hello</td>\n",
       "      <td>th@t</td>\n",
       "      <td>hello</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>whenever</td>\n",
       "      <td>th@t</td>\n",
       "      <td>@whenever/th@t</td>\n",
       "      <td>whenever</td>\n",
       "      <td>th@t</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>th@t</td>\n",
       "      <td>goodbye</td>\n",
       "      <td>@th@t/goodbye</td>\n",
       "      <td>th@t</td>\n",
       "      <td>goodbye</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cogdog</td>\n",
       "      <td>whenever</td>\n",
       "      <td>@cogdog/whenever</td>\n",
       "      <td>cogdog</td>\n",
       "      <td>whenever</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>wherever</td>\n",
       "      <td>th@t</td>\n",
       "      <td>@wherever/th@t</td>\n",
       "      <td>wherever</td>\n",
       "      <td>th@t</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       col1      col2             col3a     col3b     col3c\n",
       "0      th@t     hello       @th@t/hello      th@t     hello\n",
       "1  whenever      th@t    @whenever/th@t  whenever      th@t\n",
       "2      th@t   goodbye     @th@t/goodbye      th@t   goodbye\n",
       "3    cogdog  whenever  @cogdog/whenever    cogdog  whenever\n",
       "4  wherever      th@t    @wherever/th@t  wherever      th@t"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We could do a trivial replace, but we can also regex to be more precise\n",
    "#https://pandas.pydata.org/pandas-docs/stable/generated/pandas.Series.str.replace.html\n",
    "df['col3b'] = df['col3b'].str.replace('^@','')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}