Minimally Sufficient Guidelines.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Mininally Sufficient Guidelines\n",
    "\n",
    "This notebook contains a summary of all the guidelines in this tutorial along with a list of attributes and methods that provide nearly all of the functionality of Pandas.\n",
    "\n",
    "Minimally sufficient Pandas\n",
    "* is simple, explicit, straightforward, and boring\n",
    "* has one obvious way to accomplish a task\n",
    "* uses this obvious way every single time\n",
    "* is easier to retain in memory\n",
    "* is easier to read and debug by yourself an others\n",
    "* uses less of the library by eliminating methods that provide no additional functionality\n",
    "* avoids Pandas bugs because of less code\n",
    "* doesn't rely on being tricky to impress friends\n",
    "* makes it easier to use in production\n",
    "\n",
    "* Selecting Subsets of Data\n",
    "    * Select a single column of data with the brackets\n",
    "    * Do not use dot notation\n",
    "    * Be explicit and use `loc` and `iloc`\n",
    "    * Never use `ix`\n",
    "    * No need to use `at` or `iat`\n",
    "* Handling the `SettingWithCopyWarning`\n",
    "    * Know the three cases when it appears\n",
    "        * Correct assignment with side effects\n",
    "        * No assignment\n",
    "        * Correct assignment without side effects\n",
    "    * To handle the warning, you will be in one of two scenarios\n",
    "        * You want to work with a new independent DataFrame - use the `copy` method\n",
    "        * You want to work with original DataFrame. Assign data with a single indexer, `loc`. Avoid chained indexing.\n",
    "* Method Duplication\n",
    "    * Many methods are aliases or provide no extra functionality. Only use one\n",
    "    * All operators have methods. Only use methods when necessary \n",
    "    * Always use Pandas methods and not builtin Python functions\n",
    "* Say No to `apply`\n",
    "    * `apply` is an automated for loop that passes each column or row to a user-defined function\n",
    "    * Use `apply` as a method of last resort\n",
    "    * Using `apply` with `axis='columns'` is one of the slowest operations in all of Pandas\n",
    "* Standardizing `groupby`\n",
    "    * Know the three components\n",
    "        * Grouping columns\n",
    "        * Aggregating columns\n",
    "        * Aggregating functions\n",
    "    * Use the syntax `df.groupby('grouping columns').agg({'aggregating column': 'aggregating function'})\n",
    "* Handling a MultiIndex\n",
    "    * A MultiIndex is difficult to make selections and further process\n",
    "    * I suggest having a single level index\n",
    "    * Rename the columns and reset the index after a groupby\n",
    "* Say no to `apply` with `groupby`\n",
    "    * Can be extremely slow to use `apply` with `groupby`\n",
    "    * Call all methods independent of the group, outside of the custom function\n",
    "* Similarity between groupby, pivot_table, crosstab\n",
    "* Similarity between melt, pivot, stack, unstack\n",
    "\n",
    "## Minimal set of DataFrame attributes and methods\n",
    "Below is a short list of DataFrame attributes and methods that allows you maximum coverage of the library.\n",
    "\n",
    "* T\n",
    "* abs\n",
    "* all\n",
    "* any\n",
    "* append\n",
    "* asfreq\n",
    "* astype\n",
    "* clip\n",
    "* columns\n",
    "* copy\n",
    "* corr\n",
    "* count\n",
    "* cov\n",
    "* cummax\n",
    "* cummin\n",
    "* cumprod\n",
    "* cumsum\n",
    "* describe\n",
    "* diff\n",
    "* drop\n",
    "* drop_duplicates\n",
    "* dropna\n",
    "* dtypes\n",
    "* equals\n",
    "* expanding\n",
    "* fillna\n",
    "* groupby\n",
    "* head\n",
    "* idxmax\n",
    "* idxmin\n",
    "* iloc\n",
    "* index\n",
    "* interpolate\n",
    "* isin\n",
    "* isna\n",
    "* loc\n",
    "* max\n",
    "* mean\n",
    "* median\n",
    "* melt\n",
    "* merge\n",
    "* min\n",
    "* mode\n",
    "* nlargest\n",
    "* notna\n",
    "* nsmallest\n",
    "* nunique\n",
    "* pct_change\n",
    "* pivot_table\n",
    "* plot\n",
    "* prod\n",
    "* quantile\n",
    "* rank\n",
    "* rename\n",
    "* replace\n",
    "* resample\n",
    "* reset_index\n",
    "* rolling\n",
    "* round\n",
    "* sample\n",
    "* select_dtypes\n",
    "* shape\n",
    "* shift\n",
    "* sort_index\n",
    "* sort_values\n",
    "* std\n",
    "* sum\n",
    "* tail\n",
    "* to_csv\n",
    "* to_sql\n",
    "* values\n",
    "* var"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}