unit03_groupby.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b17f7619-3ccc-4aee-b7df-3df25631d160",
   "metadata": {},
   "source": [
    "# Unit 3: Data Manipulation - Groupby\n",
    "---\n",
    "\n",
    "1. [Working with dates](#section1)\n",
    "2. [Simple Groupby](#section2)\n",
    "3. [Sorting and filtering](#section3)\n",
    "4. [Two or more attributes](#section4)\n",
    "5. [Multiple aggregations](#section5)\n",
    "6. [Tidy up](#section6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a57b16d5-ddcb-4d4d-8fa4-1066c2aabc8d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff0411b4-aa54-4bce-b3ae-96bf483ddc7a",
   "metadata": {},
   "source": [
    "Still with airport wildlife strikes. \n",
    "[Some more info on the data](https://github.com/nlihin/data-analytics/blob/main/datasets/aircraft%20wildlife%20strikes%20fieldlist.pdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b9348d2-edea-4654-ba8a-ca69d9295f39",
   "metadata": {},
   "source": [
    "This time data from 2018-2023. This dataset was too big to upload to Github, so I split it into two parts:\\\n",
    "2018-2020\\\n",
    "2021-2023"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "40291288-61ea-4af0-ba59-bc8d23305653",
   "metadata": {},
   "outputs": [],
   "source": [
    "url1 = 'https://raw.githubusercontent.com/nlihin/data-analytics/main/datasets/aircraft%20wildlife%20strikes%202018-2020.csv'\n",
    "url2 = 'https://raw.githubusercontent.com/nlihin/data-analytics/main/datasets/aircraft%20wildlife%20strikes%202021-2023.csv'\n",
    "strike_df_18_20 = pd.read_csv(url1)\n",
    "strike_df_21_23 = pd.read_csv(url2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c898b94-931c-45e9-87e4-a40e0fd7b816",
   "metadata": {},
   "source": [
    "#### Combine the two files into one:\n",
    "Since they have the same structure, `concat` can be used.\n",
    "You will learn other options next year in DB course, but if you're curious, [look here](https://pandas.pydata.org/docs/user_guide/merging.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8ad3aa0d-d420-420c-955c-7751055e414a",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df = pd.concat([strike_df_18_20 ,strike_df_21_23]).reset_index(drop = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e90aa96-7de1-4650-9e85-951b68e28def",
   "metadata": {},
   "source": [
    "Note: `reset_index` is important here, or else you get the formar indices from both datasets (and then have multiple rows witht the same index). `drop=True` means we don't want to keep the formar index as a column"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0dd84239-2497-412e-b80d-cb5b30afd450",
   "metadata": {},
   "source": [
    "### <span style=\"color:blue\"> Exercise:</span>\n",
    "> \n",
    "> How do you suggest to check that the files have been combined? (Sanity check)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6c1cd0c9-754f-4dcd-b48b-a24e77d38ca9",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dc2ddfc8-ef00-4176-9b85-98c093e3077d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2b8f4ace-88f0-4fde-8ec8-f38b6e0e0bde",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "cfe1759a-8505-4707-8680-3d2de6039edd",
   "metadata": {
    "tags": []
   },
   "source": [
    "<a id='section1'></a>\n",
    "## 1. Working with dates"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01252a9c-b963-4867-8ff1-e53e54295106",
   "metadata": {},
   "source": [
    "We have a cell that holds the dates:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c2805014-e0f5-42d3-b2e1-5152f9d74712",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df[['INCIDENT_DATE']].tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6adc76dc-3b67-4d82-9bdd-6dc4dbb53ae1",
   "metadata": {},
   "source": [
    "But it's currently of type object:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "310be10a-65e7-4ba4-9526-96bb7c7977c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df[['INCIDENT_DATE']].dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "959a904c-c687-4488-9024-e4f654214d9f",
   "metadata": {},
   "source": [
    "We will change 'INCIDENT_DATE' into a `datetime` object "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7762add1-9969-4575-beca-abb1f76ad477",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df['date'] = pd.to_datetime(strike_df['INCIDENT_DATE'],format='%d/%m/%Y')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b0f40b5-ae1e-463b-8c7d-b353c6bddc91",
   "metadata": {},
   "source": [
    "Did it work?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "05de67cf-a5bb-4e09-adce-a85066474ca9",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df[['date']].dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0d55439e-7857-48e4-b945-cf86e0dc79a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df[['date','INCIDENT_DATE']]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "77670848-e9f9-4b47-b84e-1ca31def1619",
   "metadata": {},
   "source": [
    "Extract the month:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a36cad25-c106-4bd0-be61-0b0e1f262c6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df['month'] = pd.DatetimeIndex(strike_df['date']).month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4152d5b8-1773-4a4c-a188-320ca082a7cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df.loc[10:15,['date','month','struck_parts','SPECIES']]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27ee2a6d-1113-412b-817f-ee67881bbf1a",
   "metadata": {},
   "source": [
    "You can use any combination [from here](https://docs.python.org/3/library/datetime.html#strftime-and-strptime-behavior)\n",
    "\n",
    "Another example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ebfe30e5-c9a6-4649-9526-ea3832962255",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df['year-month'] = pd.DatetimeIndex(strike_df['date']).strftime('%Y-%m')\n",
    "strike_df[[\"year-month\",'date']].iloc[10:12]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "946999c1-8d95-4f5e-837e-bc04c1c1250e",
   "metadata": {},
   "source": [
    "### <span style=\"color:blue\"> Exercise:</span>\n",
    "> Extract the `year` and add it as a new column called `year` in `strike_df`\n",
    ">\n",
    "> Extract the name of the day and add it as a new column called `weekday` in `strike_df`\n",
    ">\n",
    "> Run the sanity check: `strike_df[[\"date\",\"year\",\"weekday\"]]` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9ad9b377-8566-47e0-aa5c-9a75c5452ac6",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "id": "853488d5-b3d9-4e4f-b6b5-8f064ccbde9a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "id": "8e1a9f53-a83a-4057-8914-b8faab0b1929",
   "metadata": {},
   "outputs": [],
   "source": [
    "# sanity check\n",
    "#strike_df[[\"date\",\"year\",\"weekday\"]]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a6cee48-44fa-40a5-ae7e-10adef4026ee",
   "metadata": {
    "tags": []
   },
   "source": [
    "<a id='section2'></a>\n",
    "## 2. Simple groupby"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f4543035-3b45-4139-931a-1df429e3351e",
   "metadata": {},
   "source": [
    "[from pandas user guide:](https://pandas.pydata.org/docs/user_guide/groupby.html)\\\n",
    "By “group by” we are referring to a process involving one or more of the following steps:\n",
    "\n",
    "* Splitting the data into groups based on some criteria.\n",
    "\n",
    "* Applying a function to each group independently.\n",
    "\n",
    "* Combining the results into a data structure.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f070909d-313b-42d1-9aa1-7933d2f48e6e",
   "metadata": {},
   "outputs": [],
   "source": [
    "grouped = strike_df.groupby('month')\n",
    "grouped"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "20093527-e2fb-4a33-a02b-11c035fc4ee0",
   "metadata": {},
   "source": [
    "The above split the data into groups\\\n",
    "Nothing happens here, since we didn't indicate what to do with each group\\\n",
    "But: no error. The split is valid :-)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc440d3b-b3ba-4382-880d-91a188ff2f75",
   "metadata": {},
   "source": [
    "Now lets perform a split and then an apply of an aggregation function"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48b82973-18af-4902-a333-16315be3de61",
   "metadata": {
    "tags": []
   },
   "source": [
    "Count the number of species that are encountered each month:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7d6b8e6e-30dc-414c-bad3-0c4e9ace614d",
   "metadata": {},
   "outputs": [],
   "source": [
    "med_df = strike_df.groupby('month')[['SPECIES']].count() \n",
    "med_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "32f16a60-407b-4c28-a425-8847e9071248",
   "metadata": {},
   "outputs": [],
   "source": [
    "#med_df[[\"month\"]]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "822b091f-452c-4acd-9029-ac442288d292",
   "metadata": {},
   "source": [
    "Note that this format means `month` is now the index\n",
    "\n",
    "this means `strike_df[[\"month\"]]` won't work anymore"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d669be06-b3e4-4ea6-9550-1a0ccd34b380",
   "metadata": {},
   "source": [
    "##### If you plan to continue using this data and need the index as an attribute:\n",
    "\n",
    "##### add `reset_index()` and then assign"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0a611934-5957-4d49-a431-f0408e1e38cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "med_df = med_df.reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ca1b340-a3fa-4633-a549-12a05467203a",
   "metadata": {},
   "source": [
    "sum some data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5ddc0407-24af-4f18-8f4b-b3892b16540e",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df[\"people_impact\"] = strike_df[['NR_INJURIES', 'NR_FATALITIES']].sum(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f134a169-3025-45d9-8cf0-978967cca26e",
   "metadata": {
    "tags": []
   },
   "source": [
    "### <span style=\"color:blue\"> Exercise:</span>\n",
    "> What is the total number of people impacted (the `people_impact`) in each month?\n",
    ">\n",
    "> What is the mean (average) of people impacted (the `people_impact`) each year?\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "433ace37-9efa-4274-abbf-27f417fa0a67",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c5cf76bd-c6f4-43ed-98a8-71ccc307b682",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "ba49a541-0bb1-4987-b34d-fd97df8a7721",
   "metadata": {},
   "source": [
    "Now let's see these on a graph:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "id": "d79e225e-e1d8-4d48-b12b-a08711e1b3bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='year', ylabel='people_impact'>"
      ]
     },
     "execution_count": 204,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "strike_df['year'] = pd.DatetimeIndex(strike_df['date']).year\n",
    "sns.barplot(data=strike_df, y ='people_impact', x = 'year',errorbar=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "id": "2771073f-be0c-4a8f-b045-98a5f2eb7ce8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='month', ylabel='people_impact'>"
      ]
     },
     "execution_count": 253,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.barplot(data=strike_df, y ='people_impact', x = 'month', estimator = sum, errorbar=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e961e0b-de6e-4e77-8a67-f50c10322b3a",
   "metadata": {},
   "source": [
    "<a id='section3'></a>\n",
    "## 3. Sorting and filtering"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "be3258ac-a470-47ca-8199-72cc52d2c959",
   "metadata": {},
   "source": [
    "Which bird species impacted people (by injuring or killing them)? How many cases were there?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0b7dad9b-2bbf-4921-8331-3b374aefab5e",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df.groupby(['SPECIES'])[['people_impact']].sum()\\\n",
    "                                                  .reset_index()\\\n",
    "                                                 .sort_values(by=\"people_impact\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33a6065b-8f2c-461a-ae93-4c3eb667b2e2",
   "metadata": {},
   "source": [
    "Most birds don't impact people. Filter out those who don't using `Filter`\\\n",
    "and then groupby again. You need to groupby twice:\n",
    "* for the filter\n",
    "* for the result after the filter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "827dc45e-af67-4283-9718-17068e9c703f",
   "metadata": {},
   "outputs": [],
   "source": [
    "bird_df = strike_df.groupby('SPECIES').filter(lambda x: x['people_impact'].sum() > 0) \\\n",
    "                            .groupby('SPECIES')[['people_impact']].sum() \\\n",
    "                            .reset_index() \\\n",
    "                            .sort_values(by=\"people_impact\", ascending = False ) \n",
    "#bird_df = bird_df.reset_index(drop = True)\n",
    "bird_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96ce8363-ead8-4005-ad04-b000e802ac55",
   "metadata": {},
   "source": [
    "another way to reach the same result, no filter:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c8fbb067-82cd-44a9-89bf-ce8708fd2606",
   "metadata": {},
   "outputs": [],
   "source": [
    "bird_df = strike_df.groupby('SPECIES')[['people_impact']].sum() \\\n",
    "                            .reset_index() \\\n",
    "                            .sort_values(by=\"people_impact\", ascending = False )\\\n",
    "                            .query('people_impact > 0')\n",
    "bird_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3df31be2-eb8a-449a-8c23-48571c1d6100",
   "metadata": {},
   "source": [
    "We cannot plot this figure without performing the groupby (and some sort of filter) first:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8818c3aa-9e26-4a4b-a795-c25e7aeb666f",
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.barplot(data = bird_df.head(6), x = \"SPECIES\", y = \"people_impact\")\n",
    "plt.xticks(rotation=45, ha='right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b4352fc-147a-4932-851c-ad3db0b50d4d",
   "metadata": {},
   "source": [
    "<a id='section4'></a>\n",
    "## 4. Groupby on two or more attributes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c145593-7302-446c-82f5-0a8a5c462798",
   "metadata": {},
   "source": [
    "Now, groupby `month` and `year`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "54d4a9ba-47ca-4bef-af7c-e9c61f286608",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_grpd = strike_df.groupby(['month','year'])[['people_impact', 'struck_parts']].sum().reset_index()\n",
    "strike_grpd"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7cc31d92-8960-43e6-b7bf-9cb962ebc63e",
   "metadata": {},
   "source": [
    "### <span style=\"color:blue\"> Exercise:</span>\n",
    "> \n",
    "> 1. what will happen if we switch the order of the indexes: `['month', 'year']`?\n",
    ">\n",
    "> 2. Create a barplot for x=\"year\", y=\"people_impact\", alongside the lineplot (see the help below)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "51810d9a-33ed-4be4-8fda-35db7644616a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "f30d2a1a-fa10-47b2-810d-31fde661fa86",
   "metadata": {},
   "source": [
    "Some help with 2:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "id": "ff50acee-e330-4c18-918f-7a3e770a3667",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(figsize=(12, 4), ncols=2)\n",
    "# write your code here\n",
    "sns.lineplot(x=\"year\", y=\"struck_parts\",  data=strike_df,errorbar=None, estimator = sum, ax=axes[1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d05bbb89-3e06-440d-ba6c-319f7cf2b909",
   "metadata": {
    "tags": []
   },
   "source": [
    "<a id='section5'></a>\n",
    "## 5. Multiple aggregations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "86f813c0-4cd5-40a5-a3c1-439d21e07be0",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df['TIME_OF_DAY'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5966a66e-647f-4914-9f34-5a8a7ed72fa1",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "strike_group = strike_df.groupby('TIME_OF_DAY').\\\n",
    "agg({'people_impact': ['mean', 'median', 'max'], 'TIME_OF_DAY':['count'], 'AircraftOutOfService':['first', 'last', 'mean']})\n",
    "\n",
    "strike_group = strike_group.reset_index()\n",
    "strike_group"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9995ac92-60d0-4889-b1d9-8b01fa2851a4",
   "metadata": {
    "tags": []
   },
   "source": [
    "<a id='section6'></a>\n",
    "## 6. Tidy your output"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1544d515-f38c-4043-80b7-4224fd1907f4",
   "metadata": {},
   "source": [
    "If you want to access the data and not deal with a multi-index, flatten the data by dropping a level and rename the columns:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f194c33a-0a2a-4e29-ba12-5bae8c0f189c",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_group.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6cba7f4c-caf4-46a6-88c0-751286bc2a9c",
   "metadata": {},
   "source": [
    "Each column currently has a multi-index, that is - several levels (two levels in our case).\n",
    "We use [droplevel](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.droplevel.html) to remove one of the indexes.\\\n",
    "`droplevel(level, axis=0)`\\\n",
    "`level` - the position of the index to drop. The topmost or leftmost index is 0.\\\n",
    "`axis` - 0 removes a level in the columns, 1 removes a level in the rows.\\\n",
    "In our case, we have two rows of index, so `axis = 1`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "22e23536-0502-4a4a-b9aa-b873e45cf25d",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_group = strike_group.droplevel(0, axis=1) \n",
    "strike_group"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cee18226-7b3f-4f96-ba1e-4e22ae3ec123",
   "metadata": {},
   "source": [
    "Rename the columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fa82845e-1875-488b-9a7e-d07a6e2ce321",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "strike_group.columns = ['time of day','impact_mean','impact_median','impact_max','time_of_day_count','AOS_first','AOS_last','AOS_mean']\n",
    "strike_group"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d2ca825d-9dbd-4ea5-a0ca-5f8ef59f6d96",
   "metadata": {},
   "source": [
    "# Save as a pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "id": "16c30c8c-dac5-446b-8be5-f307a86b921c",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df.to_pickle(\"pickled_strike\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d0388131-bba9-4d50-8504-673d1217a353",
   "metadata": {},
   "source": [
    "Read the file:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "id": "12009b16-6136-4019-9c3b-0949d4fd57a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "strike_df = pd.read_pickle(\"pickled_strike\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62ec599c-e6a8-4894-97b2-419f7fb3073a",
   "metadata": {},
   "source": [
    "Pickling the file allows us to save it with all of the changes we made in python. It is also supposed to be faster to read&write"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9c950bc-2a8b-41f4-aeb7-b7467b1c87db",
   "metadata": {},
   "source": [
    "# Summary:"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d58b0276-e3a3-4605-900f-5959d1d7c432",
   "metadata": {},
   "source": [
    "\n",
    "---\n",
    "\n",
    ">* `concat()` - combine dataframes with overlapping columns [documentation](https://pandas.pydata.org/docs/reference/api/pandas.concat.html)\n",
    ">\n",
    ">* `dtypes` - returns the data type of a series (a column in a dataframe)\n",
    ">\n",
    ">* `pd.to_datetime(df['date'])` - changes the attribute type to datetime\n",
    ">\n",
    ">* `pd.DatetimeIndex(df['date']).month` - extracts the month from the datatime attribute\n",
    ">\n",
    ">*  `pd.DatetimeIndex(strike_df['date']).strftime('%y-%m-%w')` - create a string representing the time. use %y, %a, %A etc..  \n",
    ">\n",
    ">* `groupby()` - group according to the columns specified\n",
    ">\n",
    ">* `reset_index()`  adds a numerical index\n",
    ">\n",
    ">* `apply` - applies a function on each row (axis =0) in the dataframe. Change to (axis = 1) to apply the function on each column [documentation](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.apply.html#pandas.DataFrame.apply)\n",
    ">\n",
    ">* `lambda` - small anonymous function\n",
    ">\n",
    ">* `agg` - apply multiple functions at once, one for each specified column [documentation](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.agg.html)\n",
    ">\n",
    ">* `unstack` - unstack the inner-most index onto a column\n",
    ">\n",
    ">* `droplevel(0, axis = 1)` - drops the highest (first) level in the column index of a multi-index dataframe\n",
    ">\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14eedf6a-150f-4c32-93a6-5079812cd18b",
   "metadata": {},
   "source": [
    "#### This was a lot of information.\n",
    "\n",
    "#### Keep your balance. Practice. You will make it.\n",
    "\n",
    "<div>\n",
    "<img src=\"https://raw.githubusercontent.com/nlihin/data-analytics/main/images/balance.jpg\" width=\"500\"/>\n",
    "</div>\n",
    "\n",
    "Photo by <a href=\"https://unsplash.com/@martinsanchez?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText\">Martin Sanchez</a> on <a href=\"https://unsplash.com/s/photos/perfect-balance?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText\">Unsplash</a>\n",
    "  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7a9254d1-90b5-45ee-b96b-586132b2a886",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}