ecki/Learn-Python-Programming-Third-Edition
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
2d44cd6ce545e807faf8bac0c4774c9ec858850f
Learn-Python-Programming-Th…/ch13/ch13.ipynb
T
adii1823 03bb4d43c0 ch13
2021-10-28 17:41:16 +05:30

2117 lines
239 KiB
Plaintext
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Chapter 9 - Data Science\n",
"## Data Manipulation"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 0 - Setting up the notebook"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import json\n",
"\n",
"import arrow\n",
"import numpy as np\n",
"import pandas as pd\n",
"from pandas import DataFrame"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 1- Loading Data into a DataFrame"
]
},
{
"cell_type": "code",
"execution_count": 2,
"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>cmp_name</th>\n",
" <th>cmp_bgt</th>\n",
" <th>cmp_spent</th>\n",
" <th>cmp_clicks</th>\n",
" <th>cmp_impr</th>\n",
" <th>user</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>GRZ_20210131_20210411_30-40_F_GBP</td>\n",
" <td>253951</td>\n",
" <td>17953</td>\n",
" <td>52573</td>\n",
" <td>500001</td>\n",
" <td>{\"username\": \"susan42\", \"name\": \"Emily Smith\",...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>BYU_20210109_20221204_30-35_M_GBP</td>\n",
" <td>150314</td>\n",
" <td>125884</td>\n",
" <td>24575</td>\n",
" <td>499999</td>\n",
" <td>{\"username\": \"susan42\", \"name\": \"Emily Smith\",...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>GRZ_20211124_20220921_20-35_B_EUR</td>\n",
" <td>791397</td>\n",
" <td>480963</td>\n",
" <td>39668</td>\n",
" <td>499999</td>\n",
" <td>{\"username\": \"susan42\", \"name\": \"Emily Smith\",...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>GRZ_20210727_20220211_35-45_B_EUR</td>\n",
" <td>910204</td>\n",
" <td>339997</td>\n",
" <td>16698</td>\n",
" <td>500000</td>\n",
" <td>{\"username\": \"susan42\", \"name\": \"Emily Smith\",...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>BYU_20220216_20220407_20-25_F_EUR</td>\n",
" <td>393134</td>\n",
" <td>158930</td>\n",
" <td>46631</td>\n",
" <td>500000</td>\n",
" <td>{\"username\": \"susan42\", \"name\": \"Emily Smith\",...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" cmp_name cmp_bgt cmp_spent cmp_clicks \\\n",
"0 GRZ_20210131_20210411_30-40_F_GBP 253951 17953 52573 \n",
"1 BYU_20210109_20221204_30-35_M_GBP 150314 125884 24575 \n",
"2 GRZ_20211124_20220921_20-35_B_EUR 791397 480963 39668 \n",
"3 GRZ_20210727_20220211_35-45_B_EUR 910204 339997 16698 \n",
"4 BYU_20220216_20220407_20-25_F_EUR 393134 158930 46631 \n",
"\n",
" cmp_impr user \n",
"0 500001 {\"username\": \"susan42\", \"name\": \"Emily Smith\",... \n",
"1 499999 {\"username\": \"susan42\", \"name\": \"Emily Smith\",... \n",
"2 499999 {\"username\": \"susan42\", \"name\": \"Emily Smith\",... \n",
"3 500000 {\"username\": \"susan42\", \"name\": \"Emily Smith\",... \n",
"4 500000 {\"username\": \"susan42\", \"name\": \"Emily Smith\",... "
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Load data from a json file into a DataFrame\n",
"df = pd.read_json(\"data.json\")\n",
"\n",
"# let's take a peek at the first 5 rows, to make sure\n",
"# nothing weird has happened\n",
"df.head()"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"cmp_name 5140\n",
"cmp_bgt 5140\n",
"cmp_spent 5140\n",
"cmp_clicks 5140\n",
"cmp_impr 5140\n",
"user 5140\n",
"dtype: int64"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# OK! DataFrame is alive and well!\n",
"# let's get a sense of how many rows there are and\n",
"# what is their structure.\n",
"df.count()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"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>cmp_bgt</th>\n",
" <th>cmp_spent</th>\n",
" <th>cmp_clicks</th>\n",
" <th>cmp_impr</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>count</th>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>mean</th>\n",
" <td>496331.855058</td>\n",
" <td>249542.778210</td>\n",
" <td>40414.236576</td>\n",
" <td>499999.523346</td>\n",
" </tr>\n",
" <tr>\n",
" <th>std</th>\n",
" <td>289001.241891</td>\n",
" <td>219168.636408</td>\n",
" <td>21704.136480</td>\n",
" <td>2.010877</td>\n",
" </tr>\n",
" <tr>\n",
" <th>min</th>\n",
" <td>1017.000000</td>\n",
" <td>117.000000</td>\n",
" <td>355.000000</td>\n",
" <td>499991.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25%</th>\n",
" <td>250725.500000</td>\n",
" <td>70162.000000</td>\n",
" <td>22865.250000</td>\n",
" <td>499998.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>50%</th>\n",
" <td>495957.000000</td>\n",
" <td>188704.000000</td>\n",
" <td>37103.000000</td>\n",
" <td>500000.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>75%</th>\n",
" <td>741076.500000</td>\n",
" <td>381478.750000</td>\n",
" <td>55836.000000</td>\n",
" <td>500001.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>max</th>\n",
" <td>999860.000000</td>\n",
" <td>984005.000000</td>\n",
" <td>98912.000000</td>\n",
" <td>500007.000000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" cmp_bgt cmp_spent cmp_clicks cmp_impr\n",
"count 5140.000000 5140.000000 5140.000000 5140.000000\n",
"mean 496331.855058 249542.778210 40414.236576 499999.523346\n",
"std 289001.241891 219168.636408 21704.136480 2.010877\n",
"min 1017.000000 117.000000 355.000000 499991.000000\n",
"25% 250725.500000 70162.000000 22865.250000 499998.000000\n",
"50% 495957.000000 188704.000000 37103.000000 500000.000000\n",
"75% 741076.500000 381478.750000 55836.000000 500001.000000\n",
"max 999860.000000 984005.000000 98912.000000 500007.000000"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.describe()"
]
},
{
"cell_type": "code",
"execution_count": 5,
"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>cmp_name</th>\n",
" <th>cmp_bgt</th>\n",
" <th>cmp_spent</th>\n",
" <th>cmp_clicks</th>\n",
" <th>cmp_impr</th>\n",
" <th>user</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>5047</th>\n",
" <td>GRZ_20210217_20220406_35-45_B_GBP</td>\n",
" <td>999860</td>\n",
" <td>791542</td>\n",
" <td>78932</td>\n",
" <td>499999</td>\n",
" <td>{\"username\": \"robertstephens\", \"name\": \"Jack J...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>922</th>\n",
" <td>AKX_20211111_20230908_40-50_M_GBP</td>\n",
" <td>999859</td>\n",
" <td>739683</td>\n",
" <td>73078</td>\n",
" <td>499996</td>\n",
" <td>{\"username\": \"mark20\", \"name\": \"Ronald Rojas\",...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2113</th>\n",
" <td>BYU_20220330_20220401_35-45_B_USD</td>\n",
" <td>999696</td>\n",
" <td>11791</td>\n",
" <td>42961</td>\n",
" <td>499998</td>\n",
" <td>{\"username\": \"vkennedy\", \"name\": \"Rachel Lozan...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" cmp_name cmp_bgt cmp_spent cmp_clicks \\\n",
"5047 GRZ_20210217_20220406_35-45_B_GBP 999860 791542 78932 \n",
"922 AKX_20211111_20230908_40-50_M_GBP 999859 739683 73078 \n",
"2113 BYU_20220330_20220401_35-45_B_USD 999696 11791 42961 \n",
"\n",
" cmp_impr user \n",
"5047 499999 {\"username\": \"robertstephens\", \"name\": \"Jack J... \n",
"922 499996 {\"username\": \"mark20\", \"name\": \"Ronald Rojas\",... \n",
"2113 499998 {\"username\": \"vkennedy\", \"name\": \"Rachel Lozan... "
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# let's see which are the top 3 campaigns according\n",
"# to budget (regardless of the currency)\n",
"df.sort_values(by=['cmp_bgt'], ascending=False).head(3)"
]
},
{
"cell_type": "code",
"execution_count": 6,
"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>cmp_name</th>\n",
" <th>cmp_bgt</th>\n",
" <th>cmp_spent</th>\n",
" <th>cmp_clicks</th>\n",
" <th>cmp_impr</th>\n",
" <th>user</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>2876</th>\n",
" <td>KTR_20210905_20230307_30-50_M_GBP</td>\n",
" <td>1081</td>\n",
" <td>407</td>\n",
" <td>49358</td>\n",
" <td>499999</td>\n",
" <td>{\"username\": \"kevin28\", \"name\": \"Karen Jackson...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1182</th>\n",
" <td>AKX_20210527_20220128_35-45_B_EUR</td>\n",
" <td>1054</td>\n",
" <td>392</td>\n",
" <td>36790</td>\n",
" <td>500000</td>\n",
" <td>{\"username\": \"speters\", \"name\": \"Aaron Shelton...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1954</th>\n",
" <td>BYU_20201018_20220904_25-40_F_GBP</td>\n",
" <td>1017</td>\n",
" <td>224</td>\n",
" <td>62568</td>\n",
" <td>500002</td>\n",
" <td>{\"username\": \"cdavis\", \"name\": \"Mrs. Kimberly ...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" cmp_name cmp_bgt cmp_spent cmp_clicks \\\n",
"2876 KTR_20210905_20230307_30-50_M_GBP 1081 407 49358 \n",
"1182 AKX_20210527_20220128_35-45_B_EUR 1054 392 36790 \n",
"1954 BYU_20201018_20220904_25-40_F_GBP 1017 224 62568 \n",
"\n",
" cmp_impr user \n",
"2876 499999 {\"username\": \"kevin28\", \"name\": \"Karen Jackson... \n",
"1182 500000 {\"username\": \"speters\", \"name\": \"Aaron Shelton... \n",
"1954 500002 {\"username\": \"cdavis\", \"name\": \"Mrs. Kimberly ... "
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# we can also use 'tail' to get the bottom 3 campaigns\n",
"df.sort_values(by=['cmp_bgt'], ascending=False).tail(3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 2 - Manipulating the DataFrame"
]
},
{
"cell_type": "code",
"execution_count": 7,
"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>Type</th>\n",
" <th>Start</th>\n",
" <th>End</th>\n",
" <th>Target Age</th>\n",
" <th>Target Gender</th>\n",
" <th>Currency</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>GRZ</td>\n",
" <td>2021-01-31</td>\n",
" <td>2021-04-11</td>\n",
" <td>30-40</td>\n",
" <td>F</td>\n",
" <td>GBP</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>BYU</td>\n",
" <td>2021-01-09</td>\n",
" <td>2022-12-04</td>\n",
" <td>30-35</td>\n",
" <td>M</td>\n",
" <td>GBP</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>GRZ</td>\n",
" <td>2021-11-24</td>\n",
" <td>2022-09-21</td>\n",
" <td>20-35</td>\n",
" <td>B</td>\n",
" <td>EUR</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Type Start End Target Age Target Gender Currency\n",
"0 GRZ 2021-01-31 2021-04-11 30-40 F GBP\n",
"1 BYU 2021-01-09 2022-12-04 30-35 M GBP\n",
"2 GRZ 2021-11-24 2022-09-21 20-35 B EUR"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# first, let's explode cmp_name into its components\n",
"# and get a separate DataFrame for those\n",
"\n",
"def unpack_campaign_name(name):\n",
" # very optimistic method, assumes data in campaign name\n",
" # is always in good state\n",
" type_, start, end, age, gender, currency = name.split('_')\n",
" start = arrow.get(start, 'YYYYMMDD').date()\n",
" end = arrow.get(end, 'YYYYMMDD').date()\n",
" return type_, start, end, age, gender, currency\n",
"\n",
"campaign_data = df['cmp_name'].apply(unpack_campaign_name)\n",
"campaign_cols = [\n",
" 'Type', 'Start', 'End', 'Target Age', 'Target Gender',\n",
" 'Currency']\n",
"campaign_df = DataFrame(\n",
" campaign_data.tolist(), columns=campaign_cols, index=df.index)\n",
"campaign_df.head(3)"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"# let's join the two dataframes\n",
"df = df.join(campaign_df)"
]
},
{
"cell_type": "code",
"execution_count": 9,
"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>cmp_name</th>\n",
" <th>Type</th>\n",
" <th>Start</th>\n",
" <th>End</th>\n",
" <th>Target Age</th>\n",
" <th>Target Gender</th>\n",
" <th>Currency</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>GRZ_20210131_20210411_30-40_F_GBP</td>\n",
" <td>GRZ</td>\n",
" <td>2021-01-31</td>\n",
" <td>2021-04-11</td>\n",
" <td>30-40</td>\n",
" <td>F</td>\n",
" <td>GBP</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>BYU_20210109_20221204_30-35_M_GBP</td>\n",
" <td>BYU</td>\n",
" <td>2021-01-09</td>\n",
" <td>2022-12-04</td>\n",
" <td>30-35</td>\n",
" <td>M</td>\n",
" <td>GBP</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>GRZ_20211124_20220921_20-35_B_EUR</td>\n",
" <td>GRZ</td>\n",
" <td>2021-11-24</td>\n",
" <td>2022-09-21</td>\n",
" <td>20-35</td>\n",
" <td>B</td>\n",
" <td>EUR</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" cmp_name Type Start End Target Age \\\n",
"0 GRZ_20210131_20210411_30-40_F_GBP GRZ 2021-01-31 2021-04-11 30-40 \n",
"1 BYU_20210109_20221204_30-35_M_GBP BYU 2021-01-09 2022-12-04 30-35 \n",
"2 GRZ_20211124_20220921_20-35_B_EUR GRZ 2021-11-24 2022-09-21 20-35 \n",
"\n",
" Target Gender Currency \n",
"0 F GBP \n",
"1 M GBP \n",
"2 B EUR "
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# and take a peek: good! It seems to be ok.\n",
"df[['cmp_name'] + campaign_cols].head(3)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"# now let's do the same for the JSON user object\n",
"\n",
"def unpack_user_json(user):\n",
" # very optimistic as well, expects user objects\n",
" # to have all attributes\n",
" user = json.loads(user.strip())\n",
" return [\n",
" user['username'],\n",
" user['email'],\n",
" user['name'],\n",
" user['gender'],\n",
" user['age'],\n",
" user['address'],\n",
" ]\n",
"\n",
"user_data = df['user'].apply(unpack_user_json)\n",
"user_cols = [\n",
" 'Username', 'Email', 'Name', 'Gender', 'Age', 'Address']\n",
"user_df = DataFrame(\n",
" user_data.tolist(), columns=user_cols, index=df.index)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"# let's join the two dataframes\n",
"df = df.join(user_df)"
]
},
{
"cell_type": "code",
"execution_count": 12,
"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>user</th>\n",
" <th>Username</th>\n",
" <th>Email</th>\n",
" <th>Name</th>\n",
" <th>Gender</th>\n",
" <th>Age</th>\n",
" <th>Address</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>{\"username\": \"susan42\", \"name\": \"Emily Smith\",...</td>\n",
" <td>susan42</td>\n",
" <td>vmckinney@leon.com</td>\n",
" <td>Emily Smith</td>\n",
" <td>F</td>\n",
" <td>53</td>\n",
" <td>66537 Riley Mission Apt. 337\\nNorth Jennifer, ...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>{\"username\": \"susan42\", \"name\": \"Emily Smith\",...</td>\n",
" <td>susan42</td>\n",
" <td>vmckinney@leon.com</td>\n",
" <td>Emily Smith</td>\n",
" <td>F</td>\n",
" <td>53</td>\n",
" <td>66537 Riley Mission Apt. 337\\nNorth Jennifer, ...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" user Username \\\n",
"0 {\"username\": \"susan42\", \"name\": \"Emily Smith\",... susan42 \n",
"1 {\"username\": \"susan42\", \"name\": \"Emily Smith\",... susan42 \n",
"\n",
" Email Name Gender Age \\\n",
"0 vmckinney@leon.com Emily Smith F 53 \n",
"1 vmckinney@leon.com Emily Smith F 53 \n",
"\n",
" Address \n",
"0 66537 Riley Mission Apt. 337\\nNorth Jennifer, ... \n",
"1 66537 Riley Mission Apt. 337\\nNorth Jennifer, ... "
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# and take a peek: good! Still in good shape.\n",
"df[['user'] + user_cols].head(2)"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"# now we have the DataFrame completely expanded, so it's\n",
"# time to play with it. First, let's fix those ugly column names\n",
"new_column_names = {\n",
" 'cmp_bgt': 'Budget',\n",
" 'cmp_spent': 'Spent',\n",
" 'cmp_clicks': 'Clicks',\n",
" 'cmp_impr': 'Impressions',\n",
"}\n",
"df.rename(columns=new_column_names, inplace=True)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"# let's add three other columns\n",
"\n",
"def calculate_extra_columns(df):\n",
" # Click Through Rate\n",
" df['CTR'] = df['Clicks'] / df['Impressions']\n",
" # Cost Per Click\n",
" df['CPC'] = df['Spent'] / df['Clicks']\n",
" # Cost Per Impression\n",
" df['CPI'] = df['Spent'] / df['Impressions']\n",
" \n",
"calculate_extra_columns(df)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"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>Spent</th>\n",
" <th>Clicks</th>\n",
" <th>Impressions</th>\n",
" <th>CTR</th>\n",
" <th>CPC</th>\n",
" <th>CPI</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>17953</td>\n",
" <td>52573</td>\n",
" <td>500001</td>\n",
" <td>0.105146</td>\n",
" <td>0.341487</td>\n",
" <td>0.035906</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>125884</td>\n",
" <td>24575</td>\n",
" <td>499999</td>\n",
" <td>0.049150</td>\n",
" <td>5.122442</td>\n",
" <td>0.251769</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>480963</td>\n",
" <td>39668</td>\n",
" <td>499999</td>\n",
" <td>0.079336</td>\n",
" <td>12.124710</td>\n",
" <td>0.961928</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Spent Clicks Impressions CTR CPC CPI\n",
"0 17953 52573 500001 0.105146 0.341487 0.035906\n",
"1 125884 24575 499999 0.049150 5.122442 0.251769\n",
"2 480963 39668 499999 0.079336 12.124710 0.961928"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# let's take a peek\n",
"df[['Spent', 'Clicks', 'Impressions',\n",
" 'CTR', 'CPC', 'CPI']].head(3)"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"CTR: 0.10514578970842059 0.10514578970842059\n",
"CPC: 0.3414870751146026 0.3414870751146026\n",
"CPI: 0.03590592818814362 0.03590592818814362\n"
]
}
],
"source": [
"# let's take the values of the first row and verify\n",
"clicks = df['Clicks'][0]\n",
"impressions = df['Impressions'][0]\n",
"spent = df['Spent'][0]\n",
"\n",
"CTR = df['CTR'][0]\n",
"CPC = df['CPC'][0]\n",
"CPI = df['CPI'][0]\n",
"\n",
"print('CTR:', CTR, clicks / impressions)\n",
"print('CPC:', CPC, spent / clicks)\n",
"print('CPI:', CPI, spent / impressions)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"# let's also add the name of the Day when campaign starts\n",
"def get_day_of_the_week(day):\n",
" return day.strftime(\"%A\")\n",
"\n",
"def get_duration(row):\n",
" return (row['End'] - row['Start']).days\n",
"\n",
"df['Day of Week'] = df['Start'].apply(get_day_of_the_week)\n",
"df['Duration'] = df.apply(get_duration, axis=1)"
]
},
{
"cell_type": "code",
"execution_count": 18,
"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>Start</th>\n",
" <th>End</th>\n",
" <th>Duration</th>\n",
" <th>Day of Week</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>2021-01-31</td>\n",
" <td>2021-04-11</td>\n",
" <td>70</td>\n",
" <td>Sunday</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>2021-01-09</td>\n",
" <td>2022-12-04</td>\n",
" <td>694</td>\n",
" <td>Saturday</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>2021-11-24</td>\n",
" <td>2022-09-21</td>\n",
" <td>301</td>\n",
" <td>Wednesday</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Start End Duration Day of Week\n",
"0 2021-01-31 2021-04-11 70 Sunday\n",
"1 2021-01-09 2022-12-04 694 Saturday\n",
"2 2021-11-24 2022-09-21 301 Wednesday"
]
},
"execution_count": 18,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# let's verify\n",
"df[['Start', 'End', 'Duration', 'Day of Week']].head(3)"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [],
"source": [
"# now let's get rid of the cmp_name and user columns,\n",
"# which we don't need any more, and address too\n",
"final_columns = [\n",
" 'Type', 'Start', 'End', 'Duration', 'Day of Week', 'Budget',\n",
" 'Currency', 'Clicks', 'Impressions', 'Spent', 'CTR', 'CPC',\n",
" 'CPI', 'Target Age', 'Target Gender', 'Username', 'Email',\n",
" 'Name', 'Gender', 'Age'\n",
"]\n",
"df = df[final_columns]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 3 - Saving to a file in different formats"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [],
"source": [
"# CSV format\n",
"df.to_csv('df.csv')"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
"# JSON format\n",
"df.to_json('df.json')"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [],
"source": [
"# Spreadsheet format\n",
"df.to_excel('df.xlsx')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 4 - Visualizing results"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"First let's take care of the graphics, we configure the `matplotlib` plot stle and set the font family to `serif`."
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [],
"source": [
"%matplotlib inline"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [],
"source": [
"# make the graphs nicer\n",
"import matplotlib.pyplot as plt\n",
"plt.style.use(['classic', 'ggplot'])\n",
"# see all available with: print(plt.style.available)\n",
"plt.rc('font', family='serif')"
]
},
{
"cell_type": "code",
"execution_count": 25,
"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>Duration</th>\n",
" <th>Budget</th>\n",
" <th>Clicks</th>\n",
" <th>Impressions</th>\n",
" <th>Spent</th>\n",
" <th>CTR</th>\n",
" <th>CPC</th>\n",
" <th>CPI</th>\n",
" <th>Age</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>count</th>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" <td>5140.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>mean</th>\n",
" <td>365.923930</td>\n",
" <td>496331.855058</td>\n",
" <td>40414.236576</td>\n",
" <td>499999.523346</td>\n",
" <td>249542.778210</td>\n",
" <td>0.080829</td>\n",
" <td>9.816749</td>\n",
" <td>0.499086</td>\n",
" <td>55.503891</td>\n",
" </tr>\n",
" <tr>\n",
" <th>std</th>\n",
" <td>213.233798</td>\n",
" <td>289001.241891</td>\n",
" <td>21704.136480</td>\n",
" <td>2.010877</td>\n",
" <td>219168.636408</td>\n",
" <td>0.043408</td>\n",
" <td>17.649877</td>\n",
" <td>0.438338</td>\n",
" <td>20.803059</td>\n",
" </tr>\n",
" <tr>\n",
" <th>min</th>\n",
" <td>1.000000</td>\n",
" <td>1017.000000</td>\n",
" <td>355.000000</td>\n",
" <td>499991.000000</td>\n",
" <td>117.000000</td>\n",
" <td>0.000710</td>\n",
" <td>0.003580</td>\n",
" <td>0.000234</td>\n",
" <td>18.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25%</th>\n",
" <td>180.000000</td>\n",
" <td>250725.500000</td>\n",
" <td>22865.250000</td>\n",
" <td>499998.000000</td>\n",
" <td>70162.000000</td>\n",
" <td>0.045730</td>\n",
" <td>1.778724</td>\n",
" <td>0.140325</td>\n",
" <td>38.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>50%</th>\n",
" <td>369.000000</td>\n",
" <td>495957.000000</td>\n",
" <td>37103.000000</td>\n",
" <td>500000.000000</td>\n",
" <td>188704.000000</td>\n",
" <td>0.074206</td>\n",
" <td>4.977531</td>\n",
" <td>0.377409</td>\n",
" <td>56.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>75%</th>\n",
" <td>553.000000</td>\n",
" <td>741076.500000</td>\n",
" <td>55836.000000</td>\n",
" <td>500001.000000</td>\n",
" <td>381478.750000</td>\n",
" <td>0.111673</td>\n",
" <td>11.620850</td>\n",
" <td>0.762962</td>\n",
" <td>73.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>max</th>\n",
" <td>730.000000</td>\n",
" <td>999860.000000</td>\n",
" <td>98912.000000</td>\n",
" <td>500007.000000</td>\n",
" <td>984005.000000</td>\n",
" <td>0.197824</td>\n",
" <td>517.287324</td>\n",
" <td>1.968014</td>\n",
" <td>90.000000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Duration Budget Clicks Impressions Spent \\\n",
"count 5140.000000 5140.000000 5140.000000 5140.000000 5140.000000 \n",
"mean 365.923930 496331.855058 40414.236576 499999.523346 249542.778210 \n",
"std 213.233798 289001.241891 21704.136480 2.010877 219168.636408 \n",
"min 1.000000 1017.000000 355.000000 499991.000000 117.000000 \n",
"25% 180.000000 250725.500000 22865.250000 499998.000000 70162.000000 \n",
"50% 369.000000 495957.000000 37103.000000 500000.000000 188704.000000 \n",
"75% 553.000000 741076.500000 55836.000000 500001.000000 381478.750000 \n",
"max 730.000000 999860.000000 98912.000000 500007.000000 984005.000000 \n",
"\n",
" CTR CPC CPI Age \n",
"count 5140.000000 5140.000000 5140.000000 5140.000000 \n",
"mean 0.080829 9.816749 0.499086 55.503891 \n",
"std 0.043408 17.649877 0.438338 20.803059 \n",
"min 0.000710 0.003580 0.000234 18.000000 \n",
"25% 0.045730 1.778724 0.140325 38.000000 \n",
"50% 0.074206 4.977531 0.377409 56.000000 \n",
"75% 0.111673 11.620850 0.762962 73.000000 \n",
"max 0.197824 517.287324 1.968014 90.000000 "
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.describe()"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1280x480 with 4 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"df[['Budget', 'Spent', 'Clicks', 'Impressions']].hist(\n",
" bins=16, figsize=(16, 6));"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1280x480 with 4 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"df[['CTR', 'CPC', 'CPI']].hist(\n",
" bins=20, figsize=(16, 6));"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1280x480 with 4 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# let's see the campaigns whose spent is > than 75% of the budget\n",
"selector = (df.Spent > df.Budget * .75)\n",
"df[selector][['Budget', 'Spent', 'Clicks', 'Impressions']].hist(\n",
" bins=15, figsize=(16, 6), color='green');"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {
"tags": []
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1280x480 with 3 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Let's aggregate by Day of the Week\n",
"df_weekday = df.groupby(['Day of Week']).sum()\n",
"df_weekday[['Impressions', 'Spent', 'Clicks']].plot(\n",
" figsize=(16, 6), subplots=True);"
]
},
{
"cell_type": "code",
"execution_count": 31,
"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 tr th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe thead tr:last-of-type th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr>\n",
" <th></th>\n",
" <th></th>\n",
" <th colspan=\"2\" halign=\"left\">Impressions</th>\n",
" <th colspan=\"2\" halign=\"left\">Spent</th>\n",
" </tr>\n",
" <tr>\n",
" <th></th>\n",
" <th></th>\n",
" <th>mean</th>\n",
" <th>std</th>\n",
" <th>mean</th>\n",
" <th>std</th>\n",
" </tr>\n",
" <tr>\n",
" <th>Target Gender</th>\n",
" <th>Target Age</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th rowspan=\"5\" valign=\"top\">B</th>\n",
" <th>20-25</th>\n",
" <td>499999.513514</td>\n",
" <td>2.068970</td>\n",
" <td>225522.364865</td>\n",
" <td>210082.169476</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20-30</th>\n",
" <td>499999.233766</td>\n",
" <td>2.372519</td>\n",
" <td>248960.376623</td>\n",
" <td>247992.581816</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20-35</th>\n",
" <td>499999.678161</td>\n",
" <td>1.742053</td>\n",
" <td>280387.114943</td>\n",
" <td>229581.623961</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20-40</th>\n",
" <td>499999.566265</td>\n",
" <td>1.761324</td>\n",
" <td>235929.072289</td>\n",
" <td>206375.966516</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20-45</th>\n",
" <td>499999.350000</td>\n",
" <td>2.323224</td>\n",
" <td>265572.050000</td>\n",
" <td>269311.696990</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"5\" valign=\"top\">M</th>\n",
" <th>45-50</th>\n",
" <td>499999.000000</td>\n",
" <td>1.490712</td>\n",
" <td>323302.200000</td>\n",
" <td>231532.010005</td>\n",
" </tr>\n",
" <tr>\n",
" <th>45-55</th>\n",
" <td>499999.142857</td>\n",
" <td>1.956674</td>\n",
" <td>344844.142857</td>\n",
" <td>238843.833685</td>\n",
" </tr>\n",
" <tr>\n",
" <th>45-60</th>\n",
" <td>499999.750000</td>\n",
" <td>1.332785</td>\n",
" <td>204209.750000</td>\n",
" <td>161287.282792</td>\n",
" </tr>\n",
" <tr>\n",
" <th>45-65</th>\n",
" <td>500000.125000</td>\n",
" <td>1.356203</td>\n",
" <td>249832.375000</td>\n",
" <td>190022.680442</td>\n",
" </tr>\n",
" <tr>\n",
" <th>45-70</th>\n",
" <td>500000.000000</td>\n",
" <td>0.000000</td>\n",
" <td>227598.500000</td>\n",
" <td>13319.770437</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>90 rows × 4 columns</p>\n",
"</div>"
],
"text/plain": [
" Impressions Spent \\\n",
" mean std mean \n",
"Target Gender Target Age \n",
"B 20-25 499999.513514 2.068970 225522.364865 \n",
" 20-30 499999.233766 2.372519 248960.376623 \n",
" 20-35 499999.678161 1.742053 280387.114943 \n",
" 20-40 499999.566265 1.761324 235929.072289 \n",
" 20-45 499999.350000 2.323224 265572.050000 \n",
"... ... ... ... \n",
"M 45-50 499999.000000 1.490712 323302.200000 \n",
" 45-55 499999.142857 1.956674 344844.142857 \n",
" 45-60 499999.750000 1.332785 204209.750000 \n",
" 45-65 500000.125000 1.356203 249832.375000 \n",
" 45-70 500000.000000 0.000000 227598.500000 \n",
"\n",
" \n",
" std \n",
"Target Gender Target Age \n",
"B 20-25 210082.169476 \n",
" 20-30 247992.581816 \n",
" 20-35 229581.623961 \n",
" 20-40 206375.966516 \n",
" 20-45 269311.696990 \n",
"... ... \n",
"M 45-50 231532.010005 \n",
" 45-55 238843.833685 \n",
" 45-60 161287.282792 \n",
" 45-65 190022.680442 \n",
" 45-70 13319.770437 \n",
"\n",
"[90 rows x 4 columns]"
]
},
"execution_count": 31,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Let's aggregate by gender and age\n",
"agg_config = {\n",
" 'Impressions': ['mean', 'std'],\n",
" 'Spent': ['mean', 'std'],\n",
"}\n",
"\n",
"df.groupby(['Target Gender', 'Target Age']).agg(agg_config)"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {
"tags": []
},
"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 tr th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe thead tr:last-of-type th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr>\n",
" <th></th>\n",
" <th colspan=\"3\" halign=\"left\">Clicks</th>\n",
" <th colspan=\"3\" halign=\"left\">Impressions</th>\n",
" <th colspan=\"3\" halign=\"left\">Spent</th>\n",
" </tr>\n",
" <tr>\n",
" <th>Target Gender</th>\n",
" <th>B</th>\n",
" <th>F</th>\n",
" <th>M</th>\n",
" <th>B</th>\n",
" <th>F</th>\n",
" <th>M</th>\n",
" <th>B</th>\n",
" <th>F</th>\n",
" <th>M</th>\n",
" </tr>\n",
" <tr>\n",
" <th>Target Age</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>20-25</th>\n",
" <td>2866528</td>\n",
" <td>2736143</td>\n",
" <td>3752471</td>\n",
" <td>36999964</td>\n",
" <td>35499973</td>\n",
" <td>42499984</td>\n",
" <td>16688655</td>\n",
" <td>15609550</td>\n",
" <td>22026226</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20-30</th>\n",
" <td>3377741</td>\n",
" <td>3479775</td>\n",
" <td>3297034</td>\n",
" <td>38499941</td>\n",
" <td>39999978</td>\n",
" <td>39499933</td>\n",
" <td>19169949</td>\n",
" <td>17810108</td>\n",
" <td>15777111</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20-35</th>\n",
" <td>2978342</td>\n",
" <td>2727771</td>\n",
" <td>2981045</td>\n",
" <td>43499972</td>\n",
" <td>34499979</td>\n",
" <td>38499964</td>\n",
" <td>24393679</td>\n",
" <td>18132970</td>\n",
" <td>17653304</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20-40</th>\n",
" <td>3312590</td>\n",
" <td>3269549</td>\n",
" <td>3163736</td>\n",
" <td>41499964</td>\n",
" <td>40499964</td>\n",
" <td>37999981</td>\n",
" <td>19582113</td>\n",
" <td>22317003</td>\n",
" <td>19464410</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20-45</th>\n",
" <td>760654</td>\n",
" <td>984872</td>\n",
" <td>774256</td>\n",
" <td>9999987</td>\n",
" <td>8999992</td>\n",
" <td>8499991</td>\n",
" <td>5311441</td>\n",
" <td>3461947</td>\n",
" <td>5014206</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25-30</th>\n",
" <td>2599444</td>\n",
" <td>2548985</td>\n",
" <td>3497998</td>\n",
" <td>35499970</td>\n",
" <td>35499986</td>\n",
" <td>41999977</td>\n",
" <td>16049379</td>\n",
" <td>18844103</td>\n",
" <td>19496786</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25-35</th>\n",
" <td>3576868</td>\n",
" <td>4024268</td>\n",
" <td>3019695</td>\n",
" <td>45999983</td>\n",
" <td>54999959</td>\n",
" <td>37999944</td>\n",
" <td>22770720</td>\n",
" <td>31594555</td>\n",
" <td>19591946</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25-40</th>\n",
" <td>2907057</td>\n",
" <td>2646312</td>\n",
" <td>3511864</td>\n",
" <td>36499966</td>\n",
" <td>36499976</td>\n",
" <td>41999946</td>\n",
" <td>21124388</td>\n",
" <td>17382281</td>\n",
" <td>24660000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25-45</th>\n",
" <td>2623823</td>\n",
" <td>2820536</td>\n",
" <td>2871440</td>\n",
" <td>34999991</td>\n",
" <td>35499952</td>\n",
" <td>32999963</td>\n",
" <td>20549561</td>\n",
" <td>19928836</td>\n",
" <td>16185794</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25-50</th>\n",
" <td>462439</td>\n",
" <td>753028</td>\n",
" <td>654451</td>\n",
" <td>6499999</td>\n",
" <td>9499976</td>\n",
" <td>9499977</td>\n",
" <td>3403975</td>\n",
" <td>3707541</td>\n",
" <td>3496496</td>\n",
" </tr>\n",
" <tr>\n",
" <th>30-35</th>\n",
" <td>3388423</td>\n",
" <td>2994753</td>\n",
" <td>3490426</td>\n",
" <td>43999983</td>\n",
" <td>38999956</td>\n",
" <td>41499932</td>\n",
" <td>26164157</td>\n",
" <td>15885285</td>\n",
" <td>18907029</td>\n",
" </tr>\n",
" <tr>\n",
" <th>30-40</th>\n",
" <td>3242415</td>\n",
" <td>3736380</td>\n",
" <td>3449705</td>\n",
" <td>39499954</td>\n",
" <td>44499959</td>\n",
" <td>42499939</td>\n",
" <td>19181696</td>\n",
" <td>18973465</td>\n",
" <td>18056334</td>\n",
" </tr>\n",
" <tr>\n",
" <th>30-45</th>\n",
" <td>3275455</td>\n",
" <td>2538368</td>\n",
" <td>4259140</td>\n",
" <td>36999967</td>\n",
" <td>33499981</td>\n",
" <td>49999943</td>\n",
" <td>20123897</td>\n",
" <td>20192858</td>\n",
" <td>24024862</td>\n",
" </tr>\n",
" <tr>\n",
" <th>30-50</th>\n",
" <td>2854116</td>\n",
" <td>2995629</td>\n",
" <td>3566316</td>\n",
" <td>36999929</td>\n",
" <td>36999970</td>\n",
" <td>42999951</td>\n",
" <td>17925687</td>\n",
" <td>15820300</td>\n",
" <td>21335822</td>\n",
" </tr>\n",
" <tr>\n",
" <th>30-55</th>\n",
" <td>1041929</td>\n",
" <td>607846</td>\n",
" <td>387337</td>\n",
" <td>11499978</td>\n",
" <td>7499997</td>\n",
" <td>3999986</td>\n",
" <td>5860019</td>\n",
" <td>4143309</td>\n",
" <td>2250109</td>\n",
" </tr>\n",
" <tr>\n",
" <th>35-40</th>\n",
" <td>3147055</td>\n",
" <td>3573129</td>\n",
" <td>3374785</td>\n",
" <td>38999963</td>\n",
" <td>41499979</td>\n",
" <td>41499969</td>\n",
" <td>20706965</td>\n",
" <td>17297514</td>\n",
" <td>18888529</td>\n",
" </tr>\n",
" <tr>\n",
" <th>35-45</th>\n",
" <td>3201399</td>\n",
" <td>2874475</td>\n",
" <td>2412763</td>\n",
" <td>39999945</td>\n",
" <td>36499964</td>\n",
" <td>29999965</td>\n",
" <td>20057967</td>\n",
" <td>19478538</td>\n",
" <td>12951974</td>\n",
" </tr>\n",
" <tr>\n",
" <th>35-50</th>\n",
" <td>3078118</td>\n",
" <td>3386987</td>\n",
" <td>3153942</td>\n",
" <td>38999949</td>\n",
" <td>39499974</td>\n",
" <td>42999947</td>\n",
" <td>18324889</td>\n",
" <td>18508732</td>\n",
" <td>20743245</td>\n",
" </tr>\n",
" <tr>\n",
" <th>35-55</th>\n",
" <td>3139403</td>\n",
" <td>3086026</td>\n",
" <td>3845534</td>\n",
" <td>38999981</td>\n",
" <td>39499951</td>\n",
" <td>42000006</td>\n",
" <td>19983051</td>\n",
" <td>24899403</td>\n",
" <td>20362757</td>\n",
" </tr>\n",
" <tr>\n",
" <th>35-60</th>\n",
" <td>293481</td>\n",
" <td>1122466</td>\n",
" <td>448415</td>\n",
" <td>3999983</td>\n",
" <td>11499987</td>\n",
" <td>8000000</td>\n",
" <td>1813803</td>\n",
" <td>6365702</td>\n",
" <td>5748788</td>\n",
" </tr>\n",
" <tr>\n",
" <th>40-45</th>\n",
" <td>2852606</td>\n",
" <td>2792697</td>\n",
" <td>3101402</td>\n",
" <td>33999974</td>\n",
" <td>32499975</td>\n",
" <td>36499991</td>\n",
" <td>15996311</td>\n",
" <td>16584837</td>\n",
" <td>19899182</td>\n",
" </tr>\n",
" <tr>\n",
" <th>40-50</th>\n",
" <td>3188184</td>\n",
" <td>2997715</td>\n",
" <td>3138284</td>\n",
" <td>39999955</td>\n",
" <td>37999960</td>\n",
" <td>39999983</td>\n",
" <td>17335121</td>\n",
" <td>16583301</td>\n",
" <td>23051858</td>\n",
" </tr>\n",
" <tr>\n",
" <th>40-55</th>\n",
" <td>3450353</td>\n",
" <td>3329375</td>\n",
" <td>3543599</td>\n",
" <td>39499918</td>\n",
" <td>43999982</td>\n",
" <td>40999954</td>\n",
" <td>21237290</td>\n",
" <td>23043570</td>\n",
" <td>18746124</td>\n",
" </tr>\n",
" <tr>\n",
" <th>40-60</th>\n",
" <td>3345556</td>\n",
" <td>2923719</td>\n",
" <td>3303711</td>\n",
" <td>41999951</td>\n",
" <td>35999952</td>\n",
" <td>39499956</td>\n",
" <td>19952503</td>\n",
" <td>17564038</td>\n",
" <td>19037412</td>\n",
" </tr>\n",
" <tr>\n",
" <th>40-65</th>\n",
" <td>287297</td>\n",
" <td>660921</td>\n",
" <td>250325</td>\n",
" <td>4999989</td>\n",
" <td>8000000</td>\n",
" <td>4499993</td>\n",
" <td>2527973</td>\n",
" <td>3831662</td>\n",
" <td>2226148</td>\n",
" </tr>\n",
" <tr>\n",
" <th>45-50</th>\n",
" <td>415484</td>\n",
" <td>542915</td>\n",
" <td>414364</td>\n",
" <td>4999998</td>\n",
" <td>5999983</td>\n",
" <td>4999990</td>\n",
" <td>2092140</td>\n",
" <td>4535643</td>\n",
" <td>3233022</td>\n",
" </tr>\n",
" <tr>\n",
" <th>45-55</th>\n",
" <td>686640</td>\n",
" <td>709173</td>\n",
" <td>975872</td>\n",
" <td>8999983</td>\n",
" <td>7999994</td>\n",
" <td>10499982</td>\n",
" <td>5515044</td>\n",
" <td>3716558</td>\n",
" <td>7241727</td>\n",
" </tr>\n",
" <tr>\n",
" <th>45-60</th>\n",
" <td>850162</td>\n",
" <td>718080</td>\n",
" <td>817504</td>\n",
" <td>7999995</td>\n",
" <td>9999982</td>\n",
" <td>9999995</td>\n",
" <td>3666221</td>\n",
" <td>6606485</td>\n",
" <td>4084195</td>\n",
" </tr>\n",
" <tr>\n",
" <th>45-65</th>\n",
" <td>376761</td>\n",
" <td>573359</td>\n",
" <td>297859</td>\n",
" <td>4499991</td>\n",
" <td>6999997</td>\n",
" <td>4000001</td>\n",
" <td>1838801</td>\n",
" <td>2573360</td>\n",
" <td>1998659</td>\n",
" </tr>\n",
" <tr>\n",
" <th>45-70</th>\n",
" <td>79339</td>\n",
" <td>116934</td>\n",
" <td>42055</td>\n",
" <td>1500004</td>\n",
" <td>1500002</td>\n",
" <td>1000000</td>\n",
" <td>835152</td>\n",
" <td>464627</td>\n",
" <td>455197</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Clicks Impressions \\\n",
"Target Gender B F M B F M \n",
"Target Age \n",
"20-25 2866528 2736143 3752471 36999964 35499973 42499984 \n",
"20-30 3377741 3479775 3297034 38499941 39999978 39499933 \n",
"20-35 2978342 2727771 2981045 43499972 34499979 38499964 \n",
"20-40 3312590 3269549 3163736 41499964 40499964 37999981 \n",
"20-45 760654 984872 774256 9999987 8999992 8499991 \n",
"25-30 2599444 2548985 3497998 35499970 35499986 41999977 \n",
"25-35 3576868 4024268 3019695 45999983 54999959 37999944 \n",
"25-40 2907057 2646312 3511864 36499966 36499976 41999946 \n",
"25-45 2623823 2820536 2871440 34999991 35499952 32999963 \n",
"25-50 462439 753028 654451 6499999 9499976 9499977 \n",
"30-35 3388423 2994753 3490426 43999983 38999956 41499932 \n",
"30-40 3242415 3736380 3449705 39499954 44499959 42499939 \n",
"30-45 3275455 2538368 4259140 36999967 33499981 49999943 \n",
"30-50 2854116 2995629 3566316 36999929 36999970 42999951 \n",
"30-55 1041929 607846 387337 11499978 7499997 3999986 \n",
"35-40 3147055 3573129 3374785 38999963 41499979 41499969 \n",
"35-45 3201399 2874475 2412763 39999945 36499964 29999965 \n",
"35-50 3078118 3386987 3153942 38999949 39499974 42999947 \n",
"35-55 3139403 3086026 3845534 38999981 39499951 42000006 \n",
"35-60 293481 1122466 448415 3999983 11499987 8000000 \n",
"40-45 2852606 2792697 3101402 33999974 32499975 36499991 \n",
"40-50 3188184 2997715 3138284 39999955 37999960 39999983 \n",
"40-55 3450353 3329375 3543599 39499918 43999982 40999954 \n",
"40-60 3345556 2923719 3303711 41999951 35999952 39499956 \n",
"40-65 287297 660921 250325 4999989 8000000 4499993 \n",
"45-50 415484 542915 414364 4999998 5999983 4999990 \n",
"45-55 686640 709173 975872 8999983 7999994 10499982 \n",
"45-60 850162 718080 817504 7999995 9999982 9999995 \n",
"45-65 376761 573359 297859 4499991 6999997 4000001 \n",
"45-70 79339 116934 42055 1500004 1500002 1000000 \n",
"\n",
" Spent \n",
"Target Gender B F M \n",
"Target Age \n",
"20-25 16688655 15609550 22026226 \n",
"20-30 19169949 17810108 15777111 \n",
"20-35 24393679 18132970 17653304 \n",
"20-40 19582113 22317003 19464410 \n",
"20-45 5311441 3461947 5014206 \n",
"25-30 16049379 18844103 19496786 \n",
"25-35 22770720 31594555 19591946 \n",
"25-40 21124388 17382281 24660000 \n",
"25-45 20549561 19928836 16185794 \n",
"25-50 3403975 3707541 3496496 \n",
"30-35 26164157 15885285 18907029 \n",
"30-40 19181696 18973465 18056334 \n",
"30-45 20123897 20192858 24024862 \n",
"30-50 17925687 15820300 21335822 \n",
"30-55 5860019 4143309 2250109 \n",
"35-40 20706965 17297514 18888529 \n",
"35-45 20057967 19478538 12951974 \n",
"35-50 18324889 18508732 20743245 \n",
"35-55 19983051 24899403 20362757 \n",
"35-60 1813803 6365702 5748788 \n",
"40-45 15996311 16584837 19899182 \n",
"40-50 17335121 16583301 23051858 \n",
"40-55 21237290 23043570 18746124 \n",
"40-60 19952503 17564038 19037412 \n",
"40-65 2527973 3831662 2226148 \n",
"45-50 2092140 4535643 3233022 \n",
"45-55 5515044 3716558 7241727 \n",
"45-60 3666221 6606485 4084195 \n",
"45-65 1838801 2573360 1998659 \n",
"45-70 835152 464627 455197 "
]
},
"execution_count": 32,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# finally, let's make a pivot table\n",
"df.pivot_table(\n",
" values=['Impressions', 'Clicks', 'Spent'],\n",
" index=['Target Age'],\n",
" columns=['Target Gender'],\n",
" aggfunc=np.sum\n",
")"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"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": "3.9.7"
},
"toc-autonumbering": false,
"toc-showcode": false,
"toc-showmarkdowntxt": true
},
"nbformat": 4,
"nbformat_minor": 4
}
Reference in New Issue View Git Blame Copy Permalink