nyc-taxi-gcs.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dask and cuDF on NYC Taxi CSV data on Google Cloud Storage"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Start Dask Cluster on an Eight-GPU DGX Machine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from dask.distributed import Client, wait\n",
    "from dask_cuda import LocalCUDACluster\n",
    "\n",
    "cluster = LocalCUDACluster(diagnostics_port=9000)\n",
    "client = Client(cluster)\n",
    "client"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read CSV files into Dask-GPU-DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import dask_cudf\n",
    "\n",
    "gdf = dask_cudf.read_csv('gcs://anaconda-public-data/nyc-taxi/csv/2015/yellow_*.csv')\n",
    "gdf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Time a full-pass computation\n",
    "\n",
    "Most of the time here is spent reading data from disk and parsing it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%time gdf.passenger_count.sum().compute()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Groupby Aggregations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.groupby('passenger_count').trip_distance.mean().compute().to_pandas()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### How well do people tip?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[df.tip_amount > 0]\n",
    "df = df[df.fare_amount > 0]\n",
    "df['hour'] = df.tpep_pickup_datetime.dt.hour\n",
    "hour = df.groupby('hour').trip_distance.mean().compute().to_pandas()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "hour.compute().plot(figsize=(10, 6), title='Tip Fraction by Hour')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:cudf]",
   "language": "python",
   "name": "conda-env-cudf-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}