Skip to content

Instantly share code, notes, and snippets.

@rdhyee

rdhyee/piazza_div_sum_axes.ipynb

Last active August 29, 2015 13:57
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save rdhyee/9621595 to your computer and use it in GitHub Desktop.
Save rdhyee/9621595 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
piazza_div_sum_axes.ipynb
{
"metadata": {
"name": "",
"signature": "sha256:e138f744a702941414fb2fffe3cfd420bb75352e2fbf7496bc9571d64a688ff8"
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"[Q]: (https://piazza.com/class/hqo4ux2fcns3ds?cid=10)**Why do the axis specifications behave differently for different functions?**\n",
"\n",
">Hello,\n",
" \n",
"> An example:\n",
" \n",
"```Python \n",
"df1 = DataFrame({'a': range(0,10), 'b': range(10,20)})\n",
" \n",
"df1.apply(lambda x: sum(x), axis=1) # This would sum across columns.\n",
"df1.div(df1.sum().astype(float), axis=1) # This would normalize across the rows.\n",
"df1.div(df1.sum(1).astype(float), axis=0) # This would normalize across the columns.\n",
"```\n",
" \n",
"> Thanks!\n"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from pandas import DataFrame\n",
"\n",
"df1 = DataFrame({'a': range(0,10), 'b': range(10,20)})\n",
"df1"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>a</th>\n",
" <th>b</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td> 0</td>\n",
" <td> 10</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td> 1</td>\n",
" <td> 11</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td> 2</td>\n",
" <td> 12</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td> 3</td>\n",
" <td> 13</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td> 4</td>\n",
" <td> 14</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td> 5</td>\n",
" <td> 15</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td> 6</td>\n",
" <td> 16</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td> 7</td>\n",
" <td> 17</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td> 8</td>\n",
" <td> 18</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td> 9</td>\n",
" <td> 19</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>10 rows \u00d7 2 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 1,
"text": [
" a b\n",
"0 0 10\n",
"1 1 11\n",
"2 2 12\n",
"3 3 13\n",
"4 4 14\n",
"5 5 15\n",
"6 6 16\n",
"7 7 17\n",
"8 8 18\n",
"9 9 19\n",
"\n",
"[10 rows x 2 columns]"
]
}
],
"prompt_number": 1
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df1.apply(lambda x: sum(x), axis=1) # This would sum across columns."
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 2,
"text": [
"0 10\n",
"1 12\n",
"2 14\n",
"3 16\n",
"4 18\n",
"5 20\n",
"6 22\n",
"7 24\n",
"8 26\n",
"9 28\n",
"dtype: int64"
]
}
],
"prompt_number": 2
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Useful to see what\n",
"\n",
" df1.sum().astype(float)\n",
" \n",
"does...specifically that it has index with 2 elements -- `a`, `b`"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# http://pandas.pydata.org/pandas-docs/dev/generated/pandas.DataFrame.sum.html\n",
"df1.sum().astype(float)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 3,
"text": [
"a 45\n",
"b 145\n",
"dtype: float64"
]
}
],
"prompt_number": 3
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df1.sum(1).astype(float)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 4,
"text": [
"0 10\n",
"1 12\n",
"2 14\n",
"3 16\n",
"4 18\n",
"5 20\n",
"6 22\n",
"7 24\n",
"8 26\n",
"9 28\n",
"dtype: float64"
]
}
],
"prompt_number": 4
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df1.sum(1)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 5,
"text": [
"0 10\n",
"1 12\n",
"2 14\n",
"3 16\n",
"4 18\n",
"5 20\n",
"6 22\n",
"7 24\n",
"8 26\n",
"9 28\n",
"dtype: int64"
]
}
],
"prompt_number": 5
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df1.div(df1.sum().astype(float), axis=1) # This would normalize across the rows."
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>a</th>\n",
" <th>b</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td> 0.000000</td>\n",
" <td> 0.068966</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td> 0.022222</td>\n",
" <td> 0.075862</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td> 0.044444</td>\n",
" <td> 0.082759</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td> 0.066667</td>\n",
" <td> 0.089655</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td> 0.088889</td>\n",
" <td> 0.096552</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td> 0.111111</td>\n",
" <td> 0.103448</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td> 0.133333</td>\n",
" <td> 0.110345</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td> 0.155556</td>\n",
" <td> 0.117241</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td> 0.177778</td>\n",
" <td> 0.124138</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td> 0.200000</td>\n",
" <td> 0.131034</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>10 rows \u00d7 2 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 6,
"text": [
" a b\n",
"0 0.000000 0.068966\n",
"1 0.022222 0.075862\n",
"2 0.044444 0.082759\n",
"3 0.066667 0.089655\n",
"4 0.088889 0.096552\n",
"5 0.111111 0.103448\n",
"6 0.133333 0.110345\n",
"7 0.155556 0.117241\n",
"8 0.177778 0.124138\n",
"9 0.200000 0.131034\n",
"\n",
"[10 rows x 2 columns]"
]
}
],
"prompt_number": 6
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# another way to see how df1.div(df1.sum().astype(float), axis=1) works\n",
"# axis = 1 means row by row\n",
"\n",
"for i in range(len(df1)):\n",
" print i\n",
" print df1.ix[i] / df1.sum().astype(float)\n",
" print"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"0\n",
"a 0.000000\n",
"b 0.068966\n",
"dtype: float64\n",
"\n",
"1\n",
"a 0.022222\n",
"b 0.075862\n",
"dtype: float64\n",
"\n",
"2\n",
"a 0.044444\n",
"b 0.082759\n",
"dtype: float64\n",
"\n",
"3\n",
"a 0.066667\n",
"b 0.089655\n",
"dtype: float64\n",
"\n",
"4\n",
"a 0.088889\n",
"b 0.096552\n",
"dtype: float64\n",
"\n",
"5\n",
"a 0.111111\n",
"b 0.103448\n",
"dtype: float64\n",
"\n",
"6\n",
"a 0.133333\n",
"b 0.110345\n",
"dtype: float64\n",
"\n",
"7\n",
"a 0.155556\n",
"b 0.117241\n",
"dtype: float64\n",
"\n",
"8\n",
"a 0.177778\n",
"b 0.124138\n",
"dtype: float64\n",
"\n",
"9\n",
"a 0.200000\n",
"b 0.131034\n",
"dtype: float64\n",
"\n"
]
}
],
"prompt_number": 7
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df1.div(df1.sum(1).astype(float), axis=0) # This would normalize across the columns."
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>a</th>\n",
" <th>b</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td> 0.000000</td>\n",
" <td> 1.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td> 0.083333</td>\n",
" <td> 0.916667</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td> 0.142857</td>\n",
" <td> 0.857143</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td> 0.187500</td>\n",
" <td> 0.812500</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td> 0.222222</td>\n",
" <td> 0.777778</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td> 0.250000</td>\n",
" <td> 0.750000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td> 0.272727</td>\n",
" <td> 0.727273</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td> 0.291667</td>\n",
" <td> 0.708333</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td> 0.307692</td>\n",
" <td> 0.692308</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td> 0.321429</td>\n",
" <td> 0.678571</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>10 rows \u00d7 2 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 8,
"text": [
" a b\n",
"0 0.000000 1.000000\n",
"1 0.083333 0.916667\n",
"2 0.142857 0.857143\n",
"3 0.187500 0.812500\n",
"4 0.222222 0.777778\n",
"5 0.250000 0.750000\n",
"6 0.272727 0.727273\n",
"7 0.291667 0.708333\n",
"8 0.307692 0.692308\n",
"9 0.321429 0.678571\n",
"\n",
"[10 rows x 2 columns]"
]
}
],
"prompt_number": 8
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# another way to see how df1.div(df1.sum(1).astype(float), axis=0) works\n",
"# axis=0 means column by column\n",
"\n",
"for col in df1:\n",
" print col\n",
" print df1[col]/df1.sum(1).astype(float)\n",
" print"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"a\n",
"0 0.000000\n",
"1 0.083333\n",
"2 0.142857\n",
"3 0.187500\n",
"4 0.222222\n",
"5 0.250000\n",
"6 0.272727\n",
"7 0.291667\n",
"8 0.307692\n",
"9 0.321429\n",
"dtype: float64\n",
"\n",
"b\n",
"0 1.000000\n",
"1 0.916667\n",
"2 0.857143\n",
"3 0.812500\n",
"4 0.777778\n",
"5 0.750000\n",
"6 0.727273\n",
"7 0.708333\n",
"8 0.692308\n",
"9 0.678571\n",
"dtype: float64\n",
"\n"
]
}
],
"prompt_number": 9
}
],
"metadata": {}
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment