custom-etl.ipynb

custom-futures.ipynb

timeseries.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"http://dask.readthedocs.io/en/latest/_images/dask_horizontal.svg\" \n",
    "     width=\"30%\" \n",
    "     align=right\n",
    "     alt=\"Dask logo\">\n",
    "\n",
    "Custom Workflows\n",
    "------------------\n",
    "\n",
    "We submit tasks directly to the task scheduler.  This demonstrates the flexibility that can be achieved with the `submit` function and normal Python for loops.\n",
    "\n",
    "Later on we map functions across Python queues to construct data processing pipelines."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from dask.distributed import Client, progress\n",
    "e = Client('localhost:8786')\n",
    "e"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import dask.dataframe as dd\n",
    "\n",
    "df = dd.demo.make_timeseries('2010', '2016',\n",
    "                             {'value': float, 'name': str, 'id': int},\n",
    "                             freq='10s', partition_freq='7d', seed=1)\n",
    "\n",
    "df = df[df.value > 0][['id', 'value', 'name']]\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = e.persist(df)\n",
    "progress(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%time len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%time df.groupby(df.id).value.mean().nlargest(10).compute()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Timeseries operations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df.value.resample('1d').std().head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize algorithms: Typical Groupby"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df = dd.demo.make_timeseries('2010-01-01', '2010-12-31',\n",
    "                             {'value': float, 'name': str, 'id': int},\n",
    "                             freq='10s', partition_freq='1M', seed=1)\n",
    "\n",
    "df.value.sum().visualize()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df.groupby(df.id).value.mean().visualize()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize algorithms: Datetime-resampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df = dd.demo.make_timeseries('2010-01-01', '2010-12-31',\n",
    "                             {'value': float, 'name': str, 'id': int},\n",
    "                             freq='10s', partition_freq='1M', seed=1)\n",
    "\n",
    "df.value.resample('1w').mean().visualize()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize algorithms: Rolling-aggregation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df = dd.demo.make_timeseries('2010-01-01', '2010-08-30',\n",
    "                             {'value': float, 'name': str, 'id': int},\n",
    "                             freq='10s', partition_freq='1M', seed=1)\n",
    "\n",
    "\n",
    "df.value.rolling(100).mean().visualize(rankdir='LR')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [Root]",
   "language": "python",
   "name": "Python [Root]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  },
  "widgets": {
   "state": {
    "95f143e963cc41c293bdba6e573f2c9a": {
     "views": [
      {
       "cell_index": 4
      }
     ]
    }
   },
   "version": "1.2.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}