Skip to content

Instantly share code, notes, and snippets.

@jwlin

jwlin/cs295-first-look.ipynb

Created February 14, 2018 00:40
Show Gist options
  • Save jwlin/70e6ce90d3a48ff4891c173f873a3438 to your computer and use it in GitHub Desktop.
Save jwlin/70e6ce90d3a48ff4891c173f873a3438 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
cs295-first-look.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"import os"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"col_types = {'Id': 'int', \n",
" 'OwnerUserId': 'float', \n",
" 'CreationDate': 'str', \n",
" 'ParentId': 'int', \n",
" 'Score': 'int',\n",
" 'Title': 'str',\n",
" 'Body':'str'}\n",
"\n",
"questions = pd.read_csv('/pythonquestions/Questions.csv', encoding = \"ISO-8859-1\", dtype=col_types)\n",
"answers = pd.read_csv('/pythonquestions/Answers.csv', encoding = \"ISO-8859-1\", dtype=col_types)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# A brief look of the data"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"<class 'pandas.core.frame.DataFrame'>\n",
"RangeIndex: 607282 entries, 0 to 607281\n",
"Data columns (total 6 columns):\n",
"Id 607282 non-null int32\n",
"OwnerUserId 601070 non-null float64\n",
"CreationDate 607282 non-null object\n",
"Score 607282 non-null int32\n",
"Title 607282 non-null object\n",
"Body 607282 non-null object\n",
"dtypes: float64(1), int32(2), object(3)\n",
"memory usage: 23.2+ MB\n"
]
}
],
"source": [
"questions.info()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"<class 'pandas.core.frame.DataFrame'>\n",
"RangeIndex: 987122 entries, 0 to 987121\n",
"Data columns (total 6 columns):\n",
"Id 987122 non-null int32\n",
"OwnerUserId 981755 non-null float64\n",
"CreationDate 987122 non-null object\n",
"ParentId 987122 non-null int32\n",
"Score 987122 non-null int32\n",
"Body 987122 non-null object\n",
"dtypes: float64(1), int32(3), object(2)\n",
"memory usage: 33.9+ MB\n"
]
}
],
"source": [
"answers.info()"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Id</th>\n",
" <th>OwnerUserId</th>\n",
" <th>CreationDate</th>\n",
" <th>Score</th>\n",
" <th>Title</th>\n",
" <th>Body</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>469</td>\n",
" <td>147.0</td>\n",
" <td>2008-08-02T15:11:16Z</td>\n",
" <td>21</td>\n",
" <td>How can I find the full path to a font from it...</td>\n",
" <td>&lt;p&gt;I am using the Photoshop's javascript API t...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>502</td>\n",
" <td>147.0</td>\n",
" <td>2008-08-02T17:01:58Z</td>\n",
" <td>27</td>\n",
" <td>Get a preview JPEG of a PDF on Windows?</td>\n",
" <td>&lt;p&gt;I have a cross-platform (Python) applicatio...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>535</td>\n",
" <td>154.0</td>\n",
" <td>2008-08-02T18:43:54Z</td>\n",
" <td>40</td>\n",
" <td>Continuous Integration System for a Python Cod...</td>\n",
" <td>&lt;p&gt;I'm starting work on a hobby project with a...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>594</td>\n",
" <td>116.0</td>\n",
" <td>2008-08-03T01:15:08Z</td>\n",
" <td>25</td>\n",
" <td>cx_Oracle: How do I iterate over a result set?</td>\n",
" <td>&lt;p&gt;There are several ways to iterate over a re...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>683</td>\n",
" <td>199.0</td>\n",
" <td>2008-08-03T13:19:16Z</td>\n",
" <td>28</td>\n",
" <td>Using 'in' to match an attribute of Python obj...</td>\n",
" <td>&lt;p&gt;I don't remember whether I was dreaming or ...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>742</td>\n",
" <td>189.0</td>\n",
" <td>2008-08-03T15:55:28Z</td>\n",
" <td>30</td>\n",
" <td>Class views in Django</td>\n",
" <td>&lt;p&gt;&lt;a href=\"http://www.djangoproject.com/\"&gt;Dja...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>766</td>\n",
" <td>1384652.0</td>\n",
" <td>2008-08-03T17:44:07Z</td>\n",
" <td>20</td>\n",
" <td>Python and MySQL</td>\n",
" <td>&lt;p&gt;I can get Python to work with Postgresql bu...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>773</td>\n",
" <td>207.0</td>\n",
" <td>2008-08-03T18:27:09Z</td>\n",
" <td>256</td>\n",
" <td>How do I use Python's itertools.groupby()?</td>\n",
" <td>&lt;p&gt;I haven't been able to find an understandab...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>972</td>\n",
" <td>145.0</td>\n",
" <td>2008-08-04T02:17:51Z</td>\n",
" <td>364</td>\n",
" <td>Adding a Method to an Existing Object Instance</td>\n",
" <td>&lt;p&gt;I've read that it is possible to add a meth...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td>1476</td>\n",
" <td>92.0</td>\n",
" <td>2008-08-04T18:20:36Z</td>\n",
" <td>251</td>\n",
" <td>How do you express binary literals in Python?</td>\n",
" <td>&lt;p&gt;How do you express an integer as a binary n...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Id OwnerUserId CreationDate Score \\\n",
"0 469 147.0 2008-08-02T15:11:16Z 21 \n",
"1 502 147.0 2008-08-02T17:01:58Z 27 \n",
"2 535 154.0 2008-08-02T18:43:54Z 40 \n",
"3 594 116.0 2008-08-03T01:15:08Z 25 \n",
"4 683 199.0 2008-08-03T13:19:16Z 28 \n",
"5 742 189.0 2008-08-03T15:55:28Z 30 \n",
"6 766 1384652.0 2008-08-03T17:44:07Z 20 \n",
"7 773 207.0 2008-08-03T18:27:09Z 256 \n",
"8 972 145.0 2008-08-04T02:17:51Z 364 \n",
"9 1476 92.0 2008-08-04T18:20:36Z 251 \n",
"\n",
" Title \\\n",
"0 How can I find the full path to a font from it... \n",
"1 Get a preview JPEG of a PDF on Windows? \n",
"2 Continuous Integration System for a Python Cod... \n",
"3 cx_Oracle: How do I iterate over a result set? \n",
"4 Using 'in' to match an attribute of Python obj... \n",
"5 Class views in Django \n",
"6 Python and MySQL \n",
"7 How do I use Python's itertools.groupby()? \n",
"8 Adding a Method to an Existing Object Instance \n",
"9 How do you express binary literals in Python? \n",
"\n",
" Body \n",
"0 <p>I am using the Photoshop's javascript API t... \n",
"1 <p>I have a cross-platform (Python) applicatio... \n",
"2 <p>I'm starting work on a hobby project with a... \n",
"3 <p>There are several ways to iterate over a re... \n",
"4 <p>I don't remember whether I was dreaming or ... \n",
"5 <p><a href=\"http://www.djangoproject.com/\">Dja... \n",
"6 <p>I can get Python to work with Postgresql bu... \n",
"7 <p>I haven't been able to find an understandab... \n",
"8 <p>I've read that it is possible to add a meth... \n",
"9 <p>How do you express an integer as a binary n... "
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"questions.head(10)\n",
"#questions[questions['Score']==0]"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Id</th>\n",
" <th>OwnerUserId</th>\n",
" <th>CreationDate</th>\n",
" <th>ParentId</th>\n",
" <th>Score</th>\n",
" <th>Body</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>497</td>\n",
" <td>50.0</td>\n",
" <td>2008-08-02T16:56:53Z</td>\n",
" <td>469</td>\n",
" <td>4</td>\n",
" <td>&lt;p&gt;open up a terminal (Applications-&amp;gt;Utilit...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>518</td>\n",
" <td>153.0</td>\n",
" <td>2008-08-02T17:42:28Z</td>\n",
" <td>469</td>\n",
" <td>2</td>\n",
" <td>&lt;p&gt;I haven't been able to find anything that d...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>536</td>\n",
" <td>161.0</td>\n",
" <td>2008-08-02T18:49:07Z</td>\n",
" <td>502</td>\n",
" <td>9</td>\n",
" <td>&lt;p&gt;You can use ImageMagick's convert utility f...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>538</td>\n",
" <td>156.0</td>\n",
" <td>2008-08-02T18:56:56Z</td>\n",
" <td>535</td>\n",
" <td>23</td>\n",
" <td>&lt;p&gt;One possibility is Hudson. It's written in...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>541</td>\n",
" <td>157.0</td>\n",
" <td>2008-08-02T19:06:40Z</td>\n",
" <td>535</td>\n",
" <td>20</td>\n",
" <td>&lt;p&gt;We run &lt;a href=\"http://buildbot.net/trac\"&gt;B...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>595</td>\n",
" <td>116.0</td>\n",
" <td>2008-08-03T01:17:36Z</td>\n",
" <td>594</td>\n",
" <td>25</td>\n",
" <td>&lt;p&gt;The canonical way is to use the built-in cu...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>660</td>\n",
" <td>197.0</td>\n",
" <td>2008-08-03T12:09:18Z</td>\n",
" <td>535</td>\n",
" <td>14</td>\n",
" <td>&lt;p&gt;Second the Buildbot - Trac integration. You...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>701</td>\n",
" <td>111.0</td>\n",
" <td>2008-08-03T14:30:50Z</td>\n",
" <td>683</td>\n",
" <td>3</td>\n",
" <td>&lt;p&gt;No, you were not dreaming. Python has a pr...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>735</td>\n",
" <td>145.0</td>\n",
" <td>2008-08-03T15:47:22Z</td>\n",
" <td>683</td>\n",
" <td>-2</td>\n",
" <td>&lt;p&gt;I think:&lt;/p&gt;\\r\\n\\r\\n&lt;pre&gt;&lt;code&gt;#!/bin/pytho...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td>745</td>\n",
" <td>154.0</td>\n",
" <td>2008-08-03T15:59:19Z</td>\n",
" <td>683</td>\n",
" <td>8</td>\n",
" <td>&lt;p&gt;Are you looking to get a list of objects th...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Id OwnerUserId CreationDate ParentId Score \\\n",
"0 497 50.0 2008-08-02T16:56:53Z 469 4 \n",
"1 518 153.0 2008-08-02T17:42:28Z 469 2 \n",
"2 536 161.0 2008-08-02T18:49:07Z 502 9 \n",
"3 538 156.0 2008-08-02T18:56:56Z 535 23 \n",
"4 541 157.0 2008-08-02T19:06:40Z 535 20 \n",
"5 595 116.0 2008-08-03T01:17:36Z 594 25 \n",
"6 660 197.0 2008-08-03T12:09:18Z 535 14 \n",
"7 701 111.0 2008-08-03T14:30:50Z 683 3 \n",
"8 735 145.0 2008-08-03T15:47:22Z 683 -2 \n",
"9 745 154.0 2008-08-03T15:59:19Z 683 8 \n",
"\n",
" Body \n",
"0 <p>open up a terminal (Applications-&gt;Utilit... \n",
"1 <p>I haven't been able to find anything that d... \n",
"2 <p>You can use ImageMagick's convert utility f... \n",
"3 <p>One possibility is Hudson. It's written in... \n",
"4 <p>We run <a href=\"http://buildbot.net/trac\">B... \n",
"5 <p>The canonical way is to use the built-in cu... \n",
"6 <p>Second the Buildbot - Trac integration. You... \n",
"7 <p>No, you were not dreaming. Python has a pr... \n",
"8 <p>I think:</p>\\r\\n\\r\\n<pre><code>#!/bin/pytho... \n",
"9 <p>Are you looking to get a list of objects th... "
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"answers.head(10)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# A dict from question ID to Answers (ansID, score)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"q_to_a = dict()\n",
"for ind, row in answers[['Id', 'ParentId', 'Score']].iterrows():\n",
" qid = row['ParentId']\n",
" aid = row['Id']\n",
" a_score = row['Score']\n",
" if qid not in q_to_a:\n",
" q_to_a[qid] = [(aid, a_score)]\n",
" else:\n",
" q_to_a[qid].append((aid, a_score))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# There are 539238 questions with 1+ answers. Tentitively, we gonna work on the questions w/ 4+ answers"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"539238\n",
"607282\n"
]
},
{
"data": {
"text/plain": [
"True"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"print(len(q_to_a))\n",
"print(len(questions['Id']))\n",
"set(q_to_a.keys()).issubset(set(questions['Id'].tolist()))"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"q_to_a = {k:v for k, v in q_to_a.items() if len(v) > 3}"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Statistics for questions w/ 4+ answers. For example, every question has 5.06 answers in average"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"0 4\n",
"1 7\n",
"2 8\n",
"3 9\n",
"4 6\n",
"5 9\n",
"6 16\n",
"7 6\n",
"8 4\n",
"9 4\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"count 46569.000000\n",
"mean 5.061844\n",
"std 1.990798\n",
"min 4.000000\n",
"25% 4.000000\n",
"50% 4.000000\n",
"75% 5.000000\n",
"max 55.000000\n",
"dtype: float64"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"num_ans = pd.Series([len(a) for a in q_to_a.values()])\n",
"print(num_ans[:10])\n",
"num_ans.describe()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# An instance of the worst and the best answer for a question\n",
"\n",
"We can see that the body of questions and answers are HTML, so preprocessing/data cleaning for feature extraction is required."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"982 14\n",
"984 -6\n",
"2982 576\n",
"4600 4\n",
"22525 3\n",
"959064 66\n",
"8961717 25\n",
"9041763 3\n",
"9636303 4\n",
"16240409 7\n",
"18434150 2\n",
"24748849 1\n",
"24865663 4\n",
"28060251 10\n",
"32076685 3\n",
"34404761 1\n"
]
}
],
"source": [
"for aid, score in q_to_a[972]:\n",
" print(aid, score)"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Id</th>\n",
" <th>OwnerUserId</th>\n",
" <th>CreationDate</th>\n",
" <th>Score</th>\n",
" <th>Title</th>\n",
" <th>Body</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>972</td>\n",
" <td>145.0</td>\n",
" <td>2008-08-04T02:17:51Z</td>\n",
" <td>364</td>\n",
" <td>Adding a Method to an Existing Object Instance</td>\n",
" <td>&lt;p&gt;I've read that it is possible to add a meth...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Id OwnerUserId CreationDate Score \\\n",
"8 972 145.0 2008-08-04T02:17:51Z 364 \n",
"\n",
" Title \\\n",
"8 Adding a Method to an Existing Object Instance \n",
"\n",
" Body \n",
"8 <p>I've read that it is possible to add a meth... "
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"questions[questions['Id']==972]"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<p>I've read that it is possible to add a method to an existing object (e.g. not in the class definition) in <strong>Python</strong>, I think this is called <em>Monkey Patching</em> (or in some cases <em>Duck Punching</em>). I understand that it's not always a good decision to do so. But, how might one do this?</p>\n",
"\n",
"<p><strong>UPDATE 8/04/2008 00:21:01 EST:</strong></p>\n",
"\n",
"<p><a href=\"http://stackoverflow.com/a/982\">That</a> looks like a good answer John Downey, I tried it but it appears that it ends up being not a <em>true</em> method.</p>\n",
"\n",
"<p>Your example defines the new patch function with an argument of <strong><code>self</code></strong>, but if you write actual code that way, the now patched class method asks for an argument named <code>self</code> (it doesn't automagically recognize it as the object to which it is supposed to bind, which is what would happen if defined within the class definition), meaning you have to call <strong><code>class.patch(obj)</code></strong> instead of just <strong><code>class.patch()</code></strong> if you want the same functionality as a <em>true</em> method.</p>\n",
"\n",
"<p><strong>It looks like Python isn't really treating it as a method, but more just as a variable which happens to be a function</strong> (and as such is callable). Is there any way to attach an actual method to a class?</p>\n",
"\n",
"<p>Oh, and Ryan, <a href=\"http://pypi.python.org/pypi/monkey\">that</a> isn't exactly what I was looking for (it isn't a builtin functionality), but it is quite cool nonetheless.</p>\n"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"from IPython.core.display import display, HTML\n",
"display(HTML(questions[questions['Id']==972].iloc[0]['Body']))"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Id</th>\n",
" <th>OwnerUserId</th>\n",
" <th>CreationDate</th>\n",
" <th>ParentId</th>\n",
" <th>Score</th>\n",
" <th>Body</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>16</th>\n",
" <td>984</td>\n",
" <td>122.0</td>\n",
" <td>2008-08-04T02:36:40Z</td>\n",
" <td>972</td>\n",
" <td>-6</td>\n",
" <td>&lt;p&gt;I don't know Python syntax, but I know Ruby...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Id OwnerUserId CreationDate ParentId Score \\\n",
"16 984 122.0 2008-08-04T02:36:40Z 972 -6 \n",
"\n",
" Body \n",
"16 <p>I don't know Python syntax, but I know Ruby... "
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"answers[answers['Id']==984]"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style>\n",
" .dataframe thead tr:only-child th {\n",
" text-align: right;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Id</th>\n",
" <th>OwnerUserId</th>\n",
" <th>CreationDate</th>\n",
" <th>ParentId</th>\n",
" <th>Score</th>\n",
" <th>Body</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>37</th>\n",
" <td>2982</td>\n",
" <td>99.0</td>\n",
" <td>2008-08-06T00:33:35Z</td>\n",
" <td>972</td>\n",
" <td>576</td>\n",
" <td>&lt;p&gt;In Python, there is a difference between fu...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Id OwnerUserId CreationDate ParentId Score \\\n",
"37 2982 99.0 2008-08-06T00:33:35Z 972 576 \n",
"\n",
" Body \n",
"37 <p>In Python, there is a difference between fu... "
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"answers[answers['Id']==2982]"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"<p>I don't know Python syntax, but I know Ruby can do it, and it is rather trivial. Let's say you want to add a method to Array that prints the length to standard out:</p>\r\n",
"\r\n",
"<pre><code>class Array<br> def print_length<br> puts length<br> end<br>end<br></code></pre>\r\n",
"\r\n",
"<p>If you don't want to modify the whole class, you can just add the method to a single instance of the array, and no other arrays will have the method:</p>\r\n",
"\r\n",
"<pre><code>array = [1, 2, 3]<br>def array.print_length<br> puts length<br>end<br></code></pre>\r\n",
"\r\n",
"<p>Just be aware of the issues involved in using this feature. Jeff Atwood actually <a href=\"http://www.codinghorror.com/blog/archives/001151.html\" rel=\"nofollow\">wrote about it</a> not too long ago.</p>\n"
]
}
],
"source": [
"print(answers[answers['Id']==984].iloc[0]['Body'])"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<p>I don't know Python syntax, but I know Ruby can do it, and it is rather trivial. Let's say you want to add a method to Array that prints the length to standard out:</p>\r\n",
"\r\n",
"<pre><code>class Array<br> def print_length<br> puts length<br> end<br>end<br></code></pre>\r\n",
"\r\n",
"<p>If you don't want to modify the whole class, you can just add the method to a single instance of the array, and no other arrays will have the method:</p>\r\n",
"\r\n",
"<pre><code>array = [1, 2, 3]<br>def array.print_length<br> puts length<br>end<br></code></pre>\r\n",
"\r\n",
"<p>Just be aware of the issues involved in using this feature. Jeff Atwood actually <a href=\"http://www.codinghorror.com/blog/archives/001151.html\" rel=\"nofollow\">wrote about it</a> not too long ago.</p>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"display(HTML(answers[answers['Id']==984].iloc[0]['Body']))"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"<p>In Python, there is a difference between functions and bound methods.</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; def foo():\n",
"... print \"foo\"\n",
"...\n",
"&gt;&gt;&gt; class A:\n",
"... def bar( self ):\n",
"... print \"bar\"\n",
"...\n",
"&gt;&gt;&gt; a = A()\n",
"&gt;&gt;&gt; foo\n",
"&lt;function foo at 0x00A98D70&gt;\n",
"&gt;&gt;&gt; a.bar\n",
"&lt;bound method A.bar of &lt;__main__.A instance at 0x00A9BC88&gt;&gt;\n",
"&gt;&gt;&gt;\n",
"</code></pre>\n",
"\n",
"<p>Bound methods have been \"bound\" (how descriptive) to an instance, and that instance will be passed as the first argument whenever the method is called.</p>\n",
"\n",
"<p>Callables that are attributes of a class (as opposed to an instance) are still unbound, though, so you can modify the class definition whenever you want:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; def fooFighters( self ):\n",
"... print \"fooFighters\"\n",
"...\n",
"&gt;&gt;&gt; A.fooFighters = fooFighters\n",
"&gt;&gt;&gt; a2 = A()\n",
"&gt;&gt;&gt; a2.fooFighters\n",
"&lt;bound method A.fooFighters of &lt;__main__.A instance at 0x00A9BEB8&gt;&gt;\n",
"&gt;&gt;&gt; a2.fooFighters()\n",
"fooFighters\n",
"</code></pre>\n",
"\n",
"<p>Previously defined instances are updated as well (as long as they haven't overridden the attribute themselves):</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; a.fooFighters()\n",
"fooFighters\n",
"</code></pre>\n",
"\n",
"<p>The problem comes when you want to attach a method to a single instance:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; def barFighters( self ):\n",
"... print \"barFighters\"\n",
"...\n",
"&gt;&gt;&gt; a.barFighters = barFighters\n",
"&gt;&gt;&gt; a.barFighters()\n",
"Traceback (most recent call last):\n",
" File \"&lt;stdin&gt;\", line 1, in &lt;module&gt;\n",
"TypeError: barFighters() takes exactly 1 argument (0 given)\n",
"</code></pre>\n",
"\n",
"<p>The function is not automatically bound when it's attached directly to an instance:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; a.barFighters\n",
"&lt;function barFighters at 0x00A98EF0&gt;\n",
"</code></pre>\n",
"\n",
"<p>To bind it, we can use the <a href=\"http://docs.python.org/library/types.html?highlight=methodtype#module-types\">MethodType function in the types module</a>:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; import types\n",
"&gt;&gt;&gt; a.barFighters = types.MethodType( barFighters, a )\n",
"&gt;&gt;&gt; a.barFighters\n",
"&lt;bound method ?.barFighters of &lt;__main__.A instance at 0x00A9BC88&gt;&gt;\n",
"&gt;&gt;&gt; a.barFighters()\n",
"barFighters\n",
"</code></pre>\n",
"\n",
"<p>This time other instances of the class have not been affected:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; a2.barFighters()\n",
"Traceback (most recent call last):\n",
" File \"&lt;stdin&gt;\", line 1, in &lt;module&gt;\n",
"AttributeError: A instance has no attribute 'barFighters'\n",
"</code></pre>\n",
"\n",
"<p>More information can be found by reading about <a href=\"http://users.rcn.com/python/download/Descriptor.htm\">descriptors</a> and <a href=\"http://www.onlamp.com/pub/a/python/2003/04/17/metaclasses.html\">metaclass</a> <a href=\"http://www.gnosis.cx/publish/programming/metaclass_2.html\">programming</a>.</p>\n",
"\n"
]
}
],
"source": [
"print(answers[answers['Id']==2982 ].iloc[0]['Body'])"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<p>In Python, there is a difference between functions and bound methods.</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; def foo():\n",
"... print \"foo\"\n",
"...\n",
"&gt;&gt;&gt; class A:\n",
"... def bar( self ):\n",
"... print \"bar\"\n",
"...\n",
"&gt;&gt;&gt; a = A()\n",
"&gt;&gt;&gt; foo\n",
"&lt;function foo at 0x00A98D70&gt;\n",
"&gt;&gt;&gt; a.bar\n",
"&lt;bound method A.bar of &lt;__main__.A instance at 0x00A9BC88&gt;&gt;\n",
"&gt;&gt;&gt;\n",
"</code></pre>\n",
"\n",
"<p>Bound methods have been \"bound\" (how descriptive) to an instance, and that instance will be passed as the first argument whenever the method is called.</p>\n",
"\n",
"<p>Callables that are attributes of a class (as opposed to an instance) are still unbound, though, so you can modify the class definition whenever you want:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; def fooFighters( self ):\n",
"... print \"fooFighters\"\n",
"...\n",
"&gt;&gt;&gt; A.fooFighters = fooFighters\n",
"&gt;&gt;&gt; a2 = A()\n",
"&gt;&gt;&gt; a2.fooFighters\n",
"&lt;bound method A.fooFighters of &lt;__main__.A instance at 0x00A9BEB8&gt;&gt;\n",
"&gt;&gt;&gt; a2.fooFighters()\n",
"fooFighters\n",
"</code></pre>\n",
"\n",
"<p>Previously defined instances are updated as well (as long as they haven't overridden the attribute themselves):</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; a.fooFighters()\n",
"fooFighters\n",
"</code></pre>\n",
"\n",
"<p>The problem comes when you want to attach a method to a single instance:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; def barFighters( self ):\n",
"... print \"barFighters\"\n",
"...\n",
"&gt;&gt;&gt; a.barFighters = barFighters\n",
"&gt;&gt;&gt; a.barFighters()\n",
"Traceback (most recent call last):\n",
" File \"&lt;stdin&gt;\", line 1, in &lt;module&gt;\n",
"TypeError: barFighters() takes exactly 1 argument (0 given)\n",
"</code></pre>\n",
"\n",
"<p>The function is not automatically bound when it's attached directly to an instance:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; a.barFighters\n",
"&lt;function barFighters at 0x00A98EF0&gt;\n",
"</code></pre>\n",
"\n",
"<p>To bind it, we can use the <a href=\"http://docs.python.org/library/types.html?highlight=methodtype#module-types\">MethodType function in the types module</a>:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; import types\n",
"&gt;&gt;&gt; a.barFighters = types.MethodType( barFighters, a )\n",
"&gt;&gt;&gt; a.barFighters\n",
"&lt;bound method ?.barFighters of &lt;__main__.A instance at 0x00A9BC88&gt;&gt;\n",
"&gt;&gt;&gt; a.barFighters()\n",
"barFighters\n",
"</code></pre>\n",
"\n",
"<p>This time other instances of the class have not been affected:</p>\n",
"\n",
"<pre><code>&gt;&gt;&gt; a2.barFighters()\n",
"Traceback (most recent call last):\n",
" File \"&lt;stdin&gt;\", line 1, in &lt;module&gt;\n",
"AttributeError: A instance has no attribute 'barFighters'\n",
"</code></pre>\n",
"\n",
"<p>More information can be found by reading about <a href=\"http://users.rcn.com/python/download/Descriptor.htm\">descriptors</a> and <a href=\"http://www.onlamp.com/pub/a/python/2003/04/17/metaclasses.html\">metaclass</a> <a href=\"http://www.gnosis.cx/publish/programming/metaclass_2.html\">programming</a>.</p>\n"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"display(HTML(answers[answers['Id']==2982 ].iloc[0]['Body']))"
]
}
],
"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": "3.6.3"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment