rapids.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# High Performance Jupyter\n",
    "\n",
    "## GPU time with RAPIDS\n",
    "\n",
    "<img src=\"https://rapids.ai/assets/images/RAPIDS-logo-purple.svg\" width=\"400\">\n",
    "\n",
    "We will do the same analysis as as [laptop.ipynb](laptop.ipynb), except accelerated on a GPU using RAPIDS. This notebook should work on any Linux machine with a CUDA-accelerated GPU that [RAPIDS supports](https://rapids.ai/start.html).\n",
    "\n",
    "Outputs here are from an AWS g4dn.xlarge instance (NVIDIA T4 GPU, 16GB GPU RAM)\n",
    "\n",
    "Open up a few windows from the JupyterLab NVDashboard pane on the left sidebar to monitor GPU utilization!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# cudf is the RAPIDS dataframe library (a.k.a pandas on GPU)\n",
    "import cudf  \n",
    "import numpy as np\n",
    "import datetime\n",
    "import s3fs\n",
    "import warnings\n",
    "warnings.simplefilter(\"ignore\")\n",
    "\n",
    "data_path = 's3://nyc-tlc/trip data'\n",
    "seed = 42"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load and explore data\n",
    "\n",
    "Notice that `cudf` has the same API as `pandas`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "fs = s3fs.S3FileSystem(anon=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 3.74 s, sys: 1.94 s, total: 5.68 s\n",
      "Wall time: 8.85 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "taxi = cudf.read_csv(\n",
    "    fs.open(f'{data_path}/yellow_tripdata_2019-01.csv'),\n",
    "    parse_dates=['tpep_pickup_datetime', 'tpep_dropoff_datetime'],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Row count: 7667792\n",
      "Size in GB: 1.082117204\n"
     ]
    }
   ],
   "source": [
    "print(f\"Row count: {len(taxi)}\")\n",
    "print(f\"Size in GB: {taxi.memory_usage(deep=True).sum() / 1e9}\")"
   ]
  },
  {
   "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>VendorID</th>\n",
       "      <th>tpep_pickup_datetime</th>\n",
       "      <th>tpep_dropoff_datetime</th>\n",
       "      <th>passenger_count</th>\n",
       "      <th>trip_distance</th>\n",
       "      <th>RatecodeID</th>\n",
       "      <th>store_and_fwd_flag</th>\n",
       "      <th>PULocationID</th>\n",
       "      <th>DOLocationID</th>\n",
       "      <th>payment_type</th>\n",
       "      <th>fare_amount</th>\n",
       "      <th>extra</th>\n",
       "      <th>mta_tax</th>\n",
       "      <th>tip_amount</th>\n",
       "      <th>tolls_amount</th>\n",
       "      <th>improvement_surcharge</th>\n",
       "      <th>total_amount</th>\n",
       "      <th>congestion_surcharge</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2019-01-01 00:46:40</td>\n",
       "      <td>2019-01-01 00:53:20</td>\n",
       "      <td>1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>1</td>\n",
       "      <td>N</td>\n",
       "      <td>151</td>\n",
       "      <td>239</td>\n",
       "      <td>1</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.65</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>9.95</td>\n",
       "      <td>null</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2019-01-01 00:59:47</td>\n",
       "      <td>2019-01-01 01:18:59</td>\n",
       "      <td>1</td>\n",
       "      <td>2.6</td>\n",
       "      <td>1</td>\n",
       "      <td>N</td>\n",
       "      <td>239</td>\n",
       "      <td>246</td>\n",
       "      <td>1</td>\n",
       "      <td>14.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>16.30</td>\n",
       "      <td>null</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2018-12-21 13:48:30</td>\n",
       "      <td>2018-12-21 13:52:40</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>N</td>\n",
       "      <td>236</td>\n",
       "      <td>236</td>\n",
       "      <td>1</td>\n",
       "      <td>4.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>5.80</td>\n",
       "      <td>null</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>2018-11-28 15:52:25</td>\n",
       "      <td>2018-11-28 15:55:45</td>\n",
       "      <td>5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>N</td>\n",
       "      <td>193</td>\n",
       "      <td>193</td>\n",
       "      <td>2</td>\n",
       "      <td>3.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>7.55</td>\n",
       "      <td>null</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>2018-11-28 15:56:57</td>\n",
       "      <td>2018-11-28 15:58:33</td>\n",
       "      <td>5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>N</td>\n",
       "      <td>193</td>\n",
       "      <td>193</td>\n",
       "      <td>2</td>\n",
       "      <td>52.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>55.55</td>\n",
       "      <td>null</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   VendorID tpep_pickup_datetime tpep_dropoff_datetime  passenger_count  \\\n",
       "0         1  2019-01-01 00:46:40   2019-01-01 00:53:20                1   \n",
       "1         1  2019-01-01 00:59:47   2019-01-01 01:18:59                1   \n",
       "2         2  2018-12-21 13:48:30   2018-12-21 13:52:40                3   \n",
       "3         2  2018-11-28 15:52:25   2018-11-28 15:55:45                5   \n",
       "4         2  2018-11-28 15:56:57   2018-11-28 15:58:33                5   \n",
       "\n",
       "   trip_distance  RatecodeID store_and_fwd_flag  PULocationID  DOLocationID  \\\n",
       "0            1.5           1                  N           151           239   \n",
       "1            2.6           1                  N           239           246   \n",
       "2            0.0           1                  N           236           236   \n",
       "3            0.0           1                  N           193           193   \n",
       "4            0.0           2                  N           193           193   \n",
       "\n",
       "   payment_type  fare_amount  extra  mta_tax  tip_amount  tolls_amount  \\\n",
       "0             1          7.0    0.5      0.5        1.65           0.0   \n",
       "1             1         14.0    0.5      0.5        1.00           0.0   \n",
       "2             1          4.5    0.5      0.5        0.00           0.0   \n",
       "3             2          3.5    0.5      0.5        0.00           0.0   \n",
       "4             2         52.0    0.0      0.5        0.00           0.0   \n",
       "\n",
       "   improvement_surcharge  total_amount congestion_surcharge  \n",
       "0                    0.3          9.95                 null  \n",
       "1                    0.3         16.30                 null  \n",
       "2                    0.3          5.80                 null  \n",
       "3                    0.3          7.55                 null  \n",
       "4                    0.3         55.55                 null  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "taxi.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "VendorID                          int64\n",
       "tpep_pickup_datetime     datetime64[ns]\n",
       "tpep_dropoff_datetime    datetime64[ns]\n",
       "passenger_count                   int64\n",
       "trip_distance                   float64\n",
       "RatecodeID                        int64\n",
       "store_and_fwd_flag               object\n",
       "PULocationID                      int64\n",
       "DOLocationID                      int64\n",
       "payment_type                      int64\n",
       "fare_amount                     float64\n",
       "extra                           float64\n",
       "mta_tax                         float64\n",
       "tip_amount                      float64\n",
       "tolls_amount                    float64\n",
       "improvement_surcharge           float64\n",
       "total_amount                    float64\n",
       "congestion_surcharge            float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "taxi.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 587 ms, sys: 234 ms, total: 822 ms\n",
      "Wall time: 821 ms\n"
     ]
    },
    {
     "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>VendorID</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>1.636775</td>\n",
       "      <td>0.539820</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>4.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>passenger_count</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>1.567078</td>\n",
       "      <td>1.224431</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>9.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>trip_distance</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>2.801084</td>\n",
       "      <td>3.737529</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.90</td>\n",
       "      <td>1.53</td>\n",
       "      <td>2.80</td>\n",
       "      <td>831.80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RatecodeID</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>1.058371</td>\n",
       "      <td>0.678089</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>99.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PULocationID</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>165.500918</td>\n",
       "      <td>66.391800</td>\n",
       "      <td>1.0</td>\n",
       "      <td>130.00</td>\n",
       "      <td>162.00</td>\n",
       "      <td>234.00</td>\n",
       "      <td>265.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DOLocationID</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>163.752906</td>\n",
       "      <td>70.364452</td>\n",
       "      <td>1.0</td>\n",
       "      <td>113.00</td>\n",
       "      <td>162.00</td>\n",
       "      <td>234.00</td>\n",
       "      <td>265.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>payment_type</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>1.291776</td>\n",
       "      <td>0.473323</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>4.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fare_amount</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>12.409409</td>\n",
       "      <td>262.072058</td>\n",
       "      <td>-362.0</td>\n",
       "      <td>6.00</td>\n",
       "      <td>8.50</td>\n",
       "      <td>13.50</td>\n",
       "      <td>623259.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>extra</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>0.328039</td>\n",
       "      <td>0.507479</td>\n",
       "      <td>-60.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.50</td>\n",
       "      <td>535.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mta_tax</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>0.496846</td>\n",
       "      <td>0.053378</td>\n",
       "      <td>-0.5</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.50</td>\n",
       "      <td>60.80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tip_amount</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>1.827367</td>\n",
       "      <td>2.501213</td>\n",
       "      <td>-63.5</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.43</td>\n",
       "      <td>2.33</td>\n",
       "      <td>787.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tolls_amount</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>0.316919</td>\n",
       "      <td>2.023665</td>\n",
       "      <td>-70.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>3288.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>improvement_surcharge</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>0.299338</td>\n",
       "      <td>0.019117</td>\n",
       "      <td>-0.3</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_amount</th>\n",
       "      <td>7667792.0</td>\n",
       "      <td>15.682222</td>\n",
       "      <td>262.293160</td>\n",
       "      <td>-362.8</td>\n",
       "      <td>8.19</td>\n",
       "      <td>11.27</td>\n",
       "      <td>16.56</td>\n",
       "      <td>623261.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>congestion_surcharge</th>\n",
       "      <td>2811814.0</td>\n",
       "      <td>0.000033</td>\n",
       "      <td>0.009069</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.50</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           count        mean         std    min     25%  \\\n",
       "VendorID               7667792.0    1.636775    0.539820    1.0    1.00   \n",
       "passenger_count        7667792.0    1.567078    1.224431    0.0    1.00   \n",
       "trip_distance          7667792.0    2.801084    3.737529    0.0    0.90   \n",
       "RatecodeID             7667792.0    1.058371    0.678089    1.0    1.00   \n",
       "PULocationID           7667792.0  165.500918   66.391800    1.0  130.00   \n",
       "DOLocationID           7667792.0  163.752906   70.364452    1.0  113.00   \n",
       "payment_type           7667792.0    1.291776    0.473323    1.0    1.00   \n",
       "fare_amount            7667792.0   12.409409  262.072058 -362.0    6.00   \n",
       "extra                  7667792.0    0.328039    0.507479  -60.0    0.00   \n",
       "mta_tax                7667792.0    0.496846    0.053378   -0.5    0.50   \n",
       "tip_amount             7667792.0    1.827367    2.501213  -63.5    0.00   \n",
       "tolls_amount           7667792.0    0.316919    2.023665  -70.0    0.00   \n",
       "improvement_surcharge  7667792.0    0.299338    0.019117   -0.3    0.30   \n",
       "total_amount           7667792.0   15.682222  262.293160 -362.8    8.19   \n",
       "congestion_surcharge   2811814.0    0.000033    0.009069    0.0    0.00   \n",
       "\n",
       "                          50%     75%        max  \n",
       "VendorID                 2.00    2.00       4.00  \n",
       "passenger_count          1.00    2.00       9.00  \n",
       "trip_distance            1.53    2.80     831.80  \n",
       "RatecodeID               1.00    1.00      99.00  \n",
       "PULocationID           162.00  234.00     265.00  \n",
       "DOLocationID           162.00  234.00     265.00  \n",
       "payment_type             1.00    2.00       4.00  \n",
       "fare_amount              8.50   13.50  623259.86  \n",
       "extra                    0.00    0.50     535.38  \n",
       "mta_tax                  0.50    0.50      60.80  \n",
       "tip_amount               1.43    2.33     787.25  \n",
       "tolls_amount             0.00    0.00    3288.00  \n",
       "improvement_surcharge    0.30    0.30       0.60  \n",
       "total_amount            11.27   16.56  623261.66  \n",
       "congestion_surcharge     0.00    0.00       2.50  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time \n",
    "taxi.describe().T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature engineering\n",
    "\n",
    "Same feature engineering from [laptop.ipynb](laptop.ipynb), using the same code!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "numeric_feat = [\n",
    "    'pickup_weekday', \n",
    "    'pickup_hour', \n",
    "    'pickup_week_hour', \n",
    "    'pickup_minute', \n",
    "    'passenger_count',\n",
    "]\n",
    "categorical_feat = [\n",
    "    'PULocationID', \n",
    "    'DOLocationID',\n",
    "]\n",
    "features = numeric_feat + categorical_feat\n",
    "y_col = 'high_tip'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prep_df(df: cudf.DataFrame) -> cudf.DataFrame:\n",
    "    '''\n",
    "    Generate features from a raw taxi dataframe.\n",
    "    Use 32 bit precision for GPU processing\n",
    "    '''\n",
    "    df = df[df.fare_amount > 0]  # avoid divide-by-zero\n",
    "    df['tip_fraction'] = df.tip_amount / df.fare_amount\n",
    "    df['high_tip'] = (df['tip_fraction'] > 0.2) # class label\n",
    "    \n",
    "    df['pickup_weekday'] = df.tpep_pickup_datetime.dt.weekday\n",
    "    # as of version 0.15, cudf doesn't support weekofyear\n",
    "    # df['pickup_weekofyear'] = df.tpep_pickup_datetime.dt.weekofyear\n",
    "    df['pickup_hour'] = df.tpep_pickup_datetime.dt.hour\n",
    "    df['pickup_week_hour'] = (df.pickup_weekday * 24) + df.pickup_hour\n",
    "    df['pickup_minute'] = df.tpep_pickup_datetime.dt.minute\n",
    "    df = df[features + [y_col]].astype('float32').fillna(-1)\n",
    "    df[y_col] = df[y_col].astype('int32')\n",
    "    \n",
    "    return df\n",
    "    \n",
    "taxi = prep_df(taxi)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>pickup_weekday</th>\n",
       "      <th>pickup_hour</th>\n",
       "      <th>pickup_week_hour</th>\n",
       "      <th>pickup_minute</th>\n",
       "      <th>passenger_count</th>\n",
       "      <th>PULocationID</th>\n",
       "      <th>DOLocationID</th>\n",
       "      <th>high_tip</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>151.0</td>\n",
       "      <td>239.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24.0</td>\n",
       "      <td>59.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>239.0</td>\n",
       "      <td>246.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>109.0</td>\n",
       "      <td>48.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>236.0</td>\n",
       "      <td>236.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>63.0</td>\n",
       "      <td>52.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>193.0</td>\n",
       "      <td>193.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>63.0</td>\n",
       "      <td>56.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>193.0</td>\n",
       "      <td>193.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   pickup_weekday  pickup_hour  pickup_week_hour  pickup_minute  \\\n",
       "0             1.0          0.0              24.0           46.0   \n",
       "1             1.0          0.0              24.0           59.0   \n",
       "2             4.0         13.0             109.0           48.0   \n",
       "3             2.0         15.0              63.0           52.0   \n",
       "4             2.0         15.0              63.0           56.0   \n",
       "\n",
       "   passenger_count  PULocationID  DOLocationID  high_tip  \n",
       "0              1.0         151.0         239.0         1  \n",
       "1              1.0         239.0         246.0         0  \n",
       "2              3.0         236.0         236.0         0  \n",
       "3              5.0         193.0         193.0         0  \n",
       "4              5.0         193.0         193.0         0  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "taxi.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random forest\n",
    "\n",
    "We're doing the same data splitting and training the same model as [laptop.ipynb](laptop.ipynb)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 499 ms, sys: 137 ms, total: 636 ms\n",
      "Wall time: 636 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "from cuml.preprocessing import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    taxi[features], taxi[y_col], test_size=0.33, random_state=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from cuml.ensemble import RandomForestClassifier\n",
    "rfc = RandomForestClassifier(\n",
    "    n_estimators=100, \n",
    "    max_depth=5, \n",
    "    seed=seed,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 3.19 s, sys: 1.35 s, total: 4.54 s\n",
      "Wall time: 1.31 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "_ = rfc.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 674 ms, sys: 160 ms, total: 833 ms\n",
      "Wall time: 466 ms\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array(0.5313459, dtype=float32)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "from cuml.metrics import roc_auc_score\n",
    "\n",
    "preds = rfc.predict_proba(X_test)[1]\n",
    "roc_auc_score(y_test, preds)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Nice one!\n",
    "\n",
    "This is way faster than [laptop.ipynb](laptop.ipynb), hooray for GPUs! The next logical step is: what if my data is too big for memory? \n",
    "\n",
    "Dask saves the day once again! RAPIDS uses Dask for parallelizing GPU computation across multi-node multi-GPU settings. Buckle up, and check out [rapids-dask.ipynb](rapids-dask.ipynb)."
   ]
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}