{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "DtNBZFHO3M7n" }, "source": [ "# **Automatidata project**\n", "**Course 4 - Regression Analysis: Simplify complex data relationships**" ] }, { "cell_type": "markdown", "metadata": { "id": "kaOj1equPMAb" }, "source": [ "The data consulting firm Automatidata has recently hired you as the newest member of their data analytics team. Their newest client, the NYC Taxi and Limousine Commission (New York City TLC), wants the Automatidata team to build a multiple linear regression model to predict taxi fares using existing data that was collected over the course of a year. The team is getting closer to completing the project, having completed an initial plan of action, initial Python coding work, EDA, and A/B testing.\n", "\n", "The Automatidata team has reviewed the results of the A/B testing. Now it’s time to work on predicting the taxi fare amounts. You’ve impressed your Automatidata colleagues with your hard work and attention to detail. The data team believes that you are ready to build the regression model and update the client New York City TLC about your progress.\n", "\n", "A notebook was structured and prepared to help you in this project. Please complete the following questions." ] }, { "cell_type": "markdown", "metadata": { "id": "rgSbVJvomcVa" }, "source": [ "# Course 4 End-of-course project: Build a multiple linear regression model\n", "\n", "In this activity, you will build a multiple linear regression model. As you've learned, multiple linear regression helps you estimate the linear relationship between one continuous dependent variable and two or more independent variables. For data science professionals, this is a useful skill because it allows you to consider more than one variable against the variable you're measuring against. This opens the door for much more thorough and flexible analysis to be completed. \n", "\n", "Completing this activity will help you practice planning out and buidling a multiple linear regression model based on a specific business need. The structure of this activity is designed to emulate the proposals you will likely be assigned in your career as a data professional. Completing this activity will help prepare you for those career moments.\n", "
\n", "\n", "**The purpose** of this project is to demostrate knowledge of EDA and a multiple linear regression model\n", "\n", "**The goal** is to build a multiple linear regression model and evaluate the model\n", "
\n", "*This activity has three parts:*\n", "\n", "**Part 1:** EDA & Checking Model Assumptions\n", "* What are some purposes of EDA before constructing a multiple linear regression model?\n", "\n", "**Part 2:** Model Building and evaluation\n", "* What resources do you find yourself using as you complete this stage?\n", "\n", "**Part 3:** Interpreting Model Results\n", "\n", "* What key insights emerged from your model(s)?\n", "\n", "* What business recommendations do you propose based on the models built?\n", "\n", "Follow the instructions and answer the questions below to complete the activity. Then, you will complete an Executive Summary using the questions listed on the PACE Strategy Document.\n", "\n", "Be sure to complete this activity before moving on. The next course item will provide you with a completed exemplar to compare to your own work." ] }, { "cell_type": "markdown", "metadata": { "id": "7KFOyc3JPSiN" }, "source": [ "# Build a multiple linear regression model" ] }, { "cell_type": "markdown", "metadata": { "id": "3UCHQclzQDUL" }, "source": [ "\n", "\n", "# **PACE stages**\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Throughout these project notebooks, you'll see references to the problem-solving framework PACE. The following notebook components are labeled with the respective PACE stage: Plan, Analyze, Construct, and Execute." ] }, { "cell_type": "markdown", "metadata": { "id": "F5O5cx_qQJmX" }, "source": [ "\n", "\n", "\n", "## PACE: **Plan**\n", "\n", "Consider the questions in your PACE Strategy Document to reflect on the Plan stage.\n" ] }, { "cell_type": "markdown", "metadata": { "id": "D8qYlvkLQsf2" }, "source": [ "### Task 1. Imports and loading\n", "Import the packages that you've learned are needed for building linear regression models." ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "id": "ccfeg6X6eOVZ" }, "outputs": [], "source": [ "# Imports\n", "# Packages for numerics + dataframes\n", "### YOUR CODE HERE ###\n", "import pandas as pd\n", "import numpy as np\n", "\n", "# Packages for visualization\n", "### YOUR CODE HERE ###\n", "from matplotlib import pyplot as plt\n", "import seaborn as sns\n", "\n", "# Packages for date conversions for calculating trip durations\n", "### YOUR CODE HERE ###\n", "from datetime import datetime, date\n", "\n", "# Packages for OLS, MLR, confusion matrix\n", "### YOUR CODE HERE ###\n", "from sklearn.model_selection import train_test_split\n", "from statsmodels.formula.api import ols" ] }, { "cell_type": "markdown", "metadata": { "id": "dhSYPrzQ2lpH" }, "source": [ "**Note:** `Pandas` is used to load the NYC TLC dataset. As shown in this cell, the dataset has been automatically loaded in for you. You do not need to download the .csv file, or provide more code, in order to access the dataset and proceed with this lab. Please continue with this activity by completing the following instructions." ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "id": "TyR3sBUYJBO8" }, "outputs": [], "source": [ "# Load dataset into dataframe \n", "df0=pd.read_csv(\"2017_Yellow_Taxi_Trip_Data.csv\")" ] }, { "cell_type": "markdown", "metadata": { "id": "OnrvCSfHUWPv" }, "source": [ "\n", "\n", "## PACE: **Analyze**\n", "\n", "In this stage, consider the following question where applicable to complete your code response:\n", "\n", "* What are some purposes of EDA before constructing a multiple linear regression model?\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "==> ENTER YOUR RESPONSE HERE " ] }, { "cell_type": "markdown", "metadata": { "id": "rIcDG2e66wt9" }, "source": [ "### Task 2a. Explore data with EDA\n", "\n", "Analyze and discover data, looking for correlations, missing data, outliers, and duplicates." ] }, { "cell_type": "markdown", "metadata": { "id": "CLpoUCz1277k" }, "source": [ "Start with `.shape` and `.info()`." ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "id": "T4Ag-sZhWg6K" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(22699, 18)\n", "\n", "RangeIndex: 22699 entries, 0 to 22698\n", "Data columns (total 18 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 Unnamed: 0 22699 non-null int64 \n", " 1 VendorID 22699 non-null int64 \n", " 2 tpep_pickup_datetime 22699 non-null object \n", " 3 tpep_dropoff_datetime 22699 non-null object \n", " 4 passenger_count 22699 non-null int64 \n", " 5 trip_distance 22699 non-null float64\n", " 6 RatecodeID 22699 non-null int64 \n", " 7 store_and_fwd_flag 22699 non-null object \n", " 8 PULocationID 22699 non-null int64 \n", " 9 DOLocationID 22699 non-null int64 \n", " 10 payment_type 22699 non-null int64 \n", " 11 fare_amount 22699 non-null float64\n", " 12 extra 22699 non-null float64\n", " 13 mta_tax 22699 non-null float64\n", " 14 tip_amount 22699 non-null float64\n", " 15 tolls_amount 22699 non-null float64\n", " 16 improvement_surcharge 22699 non-null float64\n", " 17 total_amount 22699 non-null float64\n", "dtypes: float64(8), int64(7), object(3)\n", "memory usage: 3.1+ MB\n", "None\n" ] } ], "source": [ "# Start with `.shape` and `.info()`\n", "### YOUR CODE HERE ###\n", "print(df0.shape)\n", "print(df0.info())" ] }, { "cell_type": "markdown", "metadata": { "id": "SWLHv_h_3Hcf" }, "source": [ "Check for missing data and duplicates using `.isna()` and `.drop_duplicates()`." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "id": "3QZZIxxi3OV3" }, "outputs": [], "source": [ "# Check for missing data and duplicates using .isna() and .drop_duplicates()\n", "### YOUR CODE HERE ###\n", "df0.drop_duplicates(inplace=True)" ] }, { "cell_type": "markdown", "metadata": { "id": "WXWAlPTY9iLK" }, "source": [ "Use `.describe()`." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "id": "2O3b9H9B9nwk" }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0VendorIDtpep_pickup_datetimetpep_dropoff_datetimepassenger_counttrip_distanceRatecodeIDstore_and_fwd_flagPULocationIDDOLocationIDpayment_typefare_amountextramta_taxtip_amounttolls_amountimprovement_surchargetotal_amount
count2.269900e+0422699.000000226992269922699.00000022699.00000022699.0000002269922699.00000022699.00000022699.00000022699.00000022699.00000022699.00000022699.00000022699.00000022699.00000022699.000000
uniqueNaNNaN2268722688NaNNaNNaN2NaNNaNNaNNaNNaNNaNNaNNaNNaNNaN
topNaNNaN07/03/2017 3:45:19 PM10/18/2017 8:07:45 PMNaNNaNNaNNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN
freqNaNNaN22NaNNaNNaN22600NaNNaNNaNNaNNaNNaNNaNNaNNaNNaN
mean5.675849e+071.556236NaNNaN1.6423192.9133131.043394NaN162.412353161.5279971.33688713.0266290.3332750.4974451.8357810.3125420.29955116.310502
std3.274493e+070.496838NaNNaN1.2852313.6531710.708391NaN66.63337370.1396910.49621113.2437910.4630970.0394652.8006261.3992120.01567316.097295
min1.212700e+041.000000NaNNaN0.0000000.0000001.000000NaN1.0000001.0000001.000000-120.000000-1.000000-0.5000000.0000000.000000-0.300000-120.300000
25%2.852056e+071.000000NaNNaN1.0000000.9900001.000000NaN114.000000112.0000001.0000006.5000000.0000000.5000000.0000000.0000000.3000008.750000
50%5.673150e+072.000000NaNNaN1.0000001.6100001.000000NaN162.000000162.0000001.0000009.5000000.0000000.5000001.3500000.0000000.30000011.800000
75%8.537452e+072.000000NaNNaN2.0000003.0600001.000000NaN233.000000233.0000002.00000014.5000000.5000000.5000002.4500000.0000000.30000017.800000
max1.134863e+082.000000NaNNaN6.00000033.96000099.000000NaN265.000000265.0000004.000000999.9900004.5000000.500000200.00000019.1000000.3000001200.290000
\n", "
" ], "text/plain": [ " Unnamed: 0 VendorID tpep_pickup_datetime \\\n", "count 2.269900e+04 22699.000000 22699 \n", "unique NaN NaN 22687 \n", "top NaN NaN 07/03/2017 3:45:19 PM \n", "freq NaN NaN 2 \n", "mean 5.675849e+07 1.556236 NaN \n", "std 3.274493e+07 0.496838 NaN \n", "min 1.212700e+04 1.000000 NaN \n", "25% 2.852056e+07 1.000000 NaN \n", "50% 5.673150e+07 2.000000 NaN \n", "75% 8.537452e+07 2.000000 NaN \n", "max 1.134863e+08 2.000000 NaN \n", "\n", " tpep_dropoff_datetime passenger_count trip_distance RatecodeID \\\n", "count 22699 22699.000000 22699.000000 22699.000000 \n", "unique 22688 NaN NaN NaN \n", "top 10/18/2017 8:07:45 PM NaN NaN NaN \n", "freq 2 NaN NaN NaN \n", "mean NaN 1.642319 2.913313 1.043394 \n", "std NaN 1.285231 3.653171 0.708391 \n", "min NaN 0.000000 0.000000 1.000000 \n", "25% NaN 1.000000 0.990000 1.000000 \n", "50% NaN 1.000000 1.610000 1.000000 \n", "75% NaN 2.000000 3.060000 1.000000 \n", "max NaN 6.000000 33.960000 99.000000 \n", "\n", " store_and_fwd_flag PULocationID DOLocationID payment_type \\\n", "count 22699 22699.000000 22699.000000 22699.000000 \n", "unique 2 NaN NaN NaN \n", "top N NaN NaN NaN \n", "freq 22600 NaN NaN NaN \n", "mean NaN 162.412353 161.527997 1.336887 \n", "std NaN 66.633373 70.139691 0.496211 \n", "min NaN 1.000000 1.000000 1.000000 \n", "25% NaN 114.000000 112.000000 1.000000 \n", "50% NaN 162.000000 162.000000 1.000000 \n", "75% NaN 233.000000 233.000000 2.000000 \n", "max NaN 265.000000 265.000000 4.000000 \n", "\n", " fare_amount extra mta_tax tip_amount tolls_amount \\\n", "count 22699.000000 22699.000000 22699.000000 22699.000000 22699.000000 \n", "unique NaN NaN NaN NaN NaN \n", "top NaN NaN NaN NaN NaN \n", "freq NaN NaN NaN NaN NaN \n", "mean 13.026629 0.333275 0.497445 1.835781 0.312542 \n", "std 13.243791 0.463097 0.039465 2.800626 1.399212 \n", "min -120.000000 -1.000000 -0.500000 0.000000 0.000000 \n", "25% 6.500000 0.000000 0.500000 0.000000 0.000000 \n", "50% 9.500000 0.000000 0.500000 1.350000 0.000000 \n", "75% 14.500000 0.500000 0.500000 2.450000 0.000000 \n", "max 999.990000 4.500000 0.500000 200.000000 19.100000 \n", "\n", " improvement_surcharge total_amount \n", "count 22699.000000 22699.000000 \n", "unique NaN NaN \n", "top NaN NaN \n", "freq NaN NaN \n", "mean 0.299551 16.310502 \n", "std 0.015673 16.097295 \n", "min -0.300000 -120.300000 \n", "25% 0.300000 8.750000 \n", "50% 0.300000 11.800000 \n", "75% 0.300000 17.800000 \n", "max 0.300000 1200.290000 " ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Use .describe()\n", "### YOUR CODE HERE ###\n", "df0.describe(include='all')" ] }, { "cell_type": "markdown", "metadata": { "id": "iXhaBfP_WOSR" }, "source": [ "### Task 2b. Convert pickup & dropoff columns to datetime\n" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "id": "TbHu-SSInJCX" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Int64Index: 22699 entries, 0 to 22698\n", "Data columns (total 2 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 tpep_pickup_datetime 22699 non-null object\n", " 1 tpep_dropoff_datetime 22699 non-null object\n", "dtypes: object(2)\n", "memory usage: 532.0+ KB\n" ] } ], "source": [ "# Check the format of the data\n", "### YOUR CODE HERE ###\n", "df0[['tpep_pickup_datetime', 'tpep_dropoff_datetime']].info()" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "id": "h5L6OdYPqV0N" }, "outputs": [], "source": [ "# Convert datetime columns to datetime\n", "### YOUR CODE HERE ###\n", "df0['tpep_pickup_datetime'] = pd.to_datetime(df0['tpep_pickup_datetime'])\n", "df0['tpep_dropoff_datetime'] = pd.to_datetime(df0['tpep_dropoff_datetime'])" ] }, { "cell_type": "markdown", "metadata": { "id": "KlF7ZNSyW0yV" }, "source": [ "### Task 2c. Create duration column" ] }, { "cell_type": "markdown", "metadata": { "id": "w1v_Y1uunbsx" }, "source": [ "Create a new column called `duration` that represents the total number of minutes that each taxi ride took." ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "id": "suC4LJFPMPCo" }, "outputs": [ { "data": { "text/plain": [ "count 22699.000000\n", "mean 17.013777\n", "std 61.996482\n", "min -16.983333\n", "25% 6.650000\n", "50% 11.183333\n", "75% 18.383333\n", "max 1439.550000\n", "Name: duration, dtype: float64" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Create `duration` column\n", "### YOUR CODE HERE ###\n", "df0['duration'] = df0['tpep_dropoff_datetime'] - df0['tpep_pickup_datetime']\n", "df0['duration'] = df0['duration'].dt.total_seconds() / 60\n", "df0['duration'].describe()" ] }, { "cell_type": "markdown", "metadata": { "id": "7dcytBKhiGAr" }, "source": [ "### Outliers\n", "\n", "Call `df.info()` to inspect the columns and decide which ones to check for outliers." ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "id": "W5bCdL5SSfg1" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Int64Index: 22699 entries, 0 to 22698\n", "Data columns (total 19 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 Unnamed: 0 22699 non-null int64 \n", " 1 VendorID 22699 non-null int64 \n", " 2 tpep_pickup_datetime 22699 non-null datetime64[ns]\n", " 3 tpep_dropoff_datetime 22699 non-null datetime64[ns]\n", " 4 passenger_count 22699 non-null int64 \n", " 5 trip_distance 22699 non-null float64 \n", " 6 RatecodeID 22699 non-null int64 \n", " 7 store_and_fwd_flag 22699 non-null object \n", " 8 PULocationID 22699 non-null int64 \n", " 9 DOLocationID 22699 non-null int64 \n", " 10 payment_type 22699 non-null int64 \n", " 11 fare_amount 22699 non-null float64 \n", " 12 extra 22699 non-null float64 \n", " 13 mta_tax 22699 non-null float64 \n", " 14 tip_amount 22699 non-null float64 \n", " 15 tolls_amount 22699 non-null float64 \n", " 16 improvement_surcharge 22699 non-null float64 \n", " 17 total_amount 22699 non-null float64 \n", " 18 duration 22699 non-null float64 \n", "dtypes: datetime64[ns](2), float64(9), int64(7), object(1)\n", "memory usage: 3.5+ MB\n" ] } ], "source": [ "### YOUR CODE HERE ###\n", "df0.info()\n", "# trip_distance, fare_amount, total_amount, duration" ] }, { "cell_type": "markdown", "metadata": { "id": "gS7VR2S0izZE" }, "source": [ "Keeping in mind that many of the features will not be used to fit your model, the most important columns to check for outliers are likely to be:\n", "* `trip_distance`\n", "* `fare_amount`\n", "* `duration`\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "id": "Vtj4iAJMk9Vc" }, "source": [ "### Task 2d. Box plots\n", "\n", "Plot a box plot for each feature: `trip_distance`, `fare_amount`, `duration`." ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "id": "KCEzE-gwL5gq" }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAEHCAYAAAB/UveQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAQu0lEQVR4nO3df2zc9X3H8dfbPpdQoMMYFkUO5dLeJlaxia0R+0E3hYlMdtjUdVK7VWJxB1NbbQ0eSsXK5ihhHFIVtnXBrYqYhogHXduV0RVhew1T6I9toiSUEOqkcMqchZTxw9CWgEft5L0/7nunu/Pd2Wff3fuSPB9SlLvP9/P9ft7fj+yXv/7c3dfm7gIAxOiKLgAAzmaEMAAEIoQBIBAhDACBCGEACJRqpPPFF1/s6XS6RaUAwJlp//79r7j7JdW2NRTC6XRa+/bta05VAHCWMLOjtbaxHAEAgQhhAAhECANAIEIYAAIRwgAQiBAGgECEMAAEIoQBIBAhDACBCGEACEQIA0AgQhgAAhHCABCIEAaAQIQwAAQihAEgECEMAIEIYQAIRAgDQKCG/sZcK42OjiqXy5W1HT9+XJLU39/f0LEymYy2bNnStNoAoFU6JoRzuZyeeuaQTr79omJb95s/kiT971tLL7P7zVebXhsAtErHhLAknXz7RZq9fFPx+bmHxyWprG0xhX0A4HTAmjAABCKEASAQIQwAgQhhAAhECANAIEIYAAIRwgAQiBAGgECEMAAEIoQBIBAhDACBCGEACEQIA0AgQhgAAhHCABCIEAaAQIQwAAQihAEgECEMAIEIYQAIRAgDQCBCGAACEcIAEIgQBoBAhDAABCKEASAQIQwAgQhhAAhECANAIEIYAAIRwgAQiBAGgECEMAAEIoQBIBAhDACBCGEACEQIA0AgQhgAAhHCABCIEAaAQIQwAARqSwiPjo5qdHS0HUO13Zl8bgBaL9WOQXK5XDuGCXEmnxuA1mM5AgACEcIAEIgQBoBAhDAABCKEASAQIQwAgQhhAAhECANAIEIYAAIRwgAQiBAGgECEMAAEIoQBIBAhDACBCGEACEQIA0AgQhgAAhHCABCIEAaAQIQwAAQihAEgECEMAIEIYQAIRAgDQCBCGAACEcIAEIgQBoBAhDAABCKEASAQIQwAgQhhAAhECANAIEIYAAIRwgAQiBAGgECEMAAEIoQBIBAhDACBUtEFnO4OHDggSdqwYUNsIWeIrq4unTp1KrqMRXV3d+vkyZNVt9U6h97eXr322muSpFQq/603Pz+vVColM9Pc3Jx6enokSXNzc7r++uv1wAMPaN26dbrzzjvV19enmZkZbdu2Te6ubDa7oG3r1q3auXOnjh07ptHRUfX29uq2227T9u3bi31vvfVWHTt2TLfccot27typNWvWaNWqVbrhhhu0bds2rVmzRuecc462bt2qu+66S9u3b5ek4nEkLahBUlkdN954o7Zv365du3Ypk8msaK5nZmY0MjIiM9Ptt99eVkth7Fr71epXb1ulXC6n4eHhppxLNVwJo6OcDgEsqWYAS7XPoRDAUj585+fni4/n5uYk5cO38Pj++++Xu+vIkSMaGxuTJO3evVtTU1M6dOhQ1bZsNqtnn31Ws7Ozymaz2r17tw4ePFjWt7D9jjvu0OzsrI4cOaKpqSnt2LGj+LxwrMK+pcepVkNlHTt27NAbb7yhbDa73CkuO+6hQ4c0NTW1oJbF9qvVb6nHkKRsNtu0c6mGEF6Ba665JroEnCUeeeQR5XI5TU5OFtsmJiYWtE1PT5c9Hh8fl7trcnJSuVxOExMTxe2FHwIFJ06cKHs+PT0td9fExIQmJibKHpfWMDMzo5mZmbI6Cseanp5WLpdb9nnPzMyUjTc+Pl6sZXJyUjMzMzX3m5ycrNqv3rZKuVyuOKcrPZda2rIccfz4cc3Ozmp4eLhmn1wup66f+IrH6vq/HyuXe73uWM3ivvJ6gaWYn59XNpstXiVL+avmyrZq+0n5K/fF+tZSOWbp1/3c3JzGxsbk7jWPnc1mdd999zU8rpS/Yi39YTE3Nyczk5Q/p7GxMd18881V9yv8RlLZr962arU361xqWfRK2Mw+amb7zGzfyy+/3NTBASxd4cq0wN0XtNUyPz9fdpXcCHcvjlE5lrtrz549evTRR2vWsdxxJVU9buH5/Py89uzZU3O/0uWe0n71ti1W+0rOpZZFr4Td/R5J90jS+vXrl3Xp19/fL0natWtXzT7Dw8Paf+TF5Ry+zKlV71DmXavrjtUsvBiHdkqn0zp69GgxhMxMl112WVlbLalUSmvXrl1WiBSuPN1dZlY2lplp48aNcnc9/PDDVetIp9MNj1lw7bXXLjhuoYZUKqWNGzfW3G98fLz4wmdpv3rbqtVeOmcrOZdaWBNegcIXJ9BqqVRKIyMjxXdPSFJPT8+Ctmr7Sfl3cyzWt5aenp7icXp6ehbUsHnzZg0NDdU89sjISMNjFgwNDRXHrqylu7tbmzdvrrlfV1dX1X71ti1W+0rOpRZCeAX27t0bXQLOEtddd50ymYwGBgaKbYODgwvaSq/U0um0Nm3aJDPTwMCAMpmMBgcHi9tLw02Szj///LLn6XRaZqbBwUENDg6WPS6toa+vT319fWV1FI6VTqdX9Lauvr6+svE2bdpUrGVgYKDm28sK9VTrV29bpUwmU5zTlZ5LLbxPGB2F9wlXf59w4WptaGhIuVxO7l61rfR9wiMjI+rt7dX09HRZ38OHDzf0PuHCvqXHqayhso7C+4SbceU4NDSk5557TmZWtZZ6+9XqV29bpZGREQ0PD7fkKliSrJFX+NevX+/79u1reJDCOxWWsiY8e/mmYtu5h8clqaxtMeceHtd727QmLC3t3ACc3cxsv7uvr7aN5QgACEQIA0AgQhgAAhHCABCIEAaAQIQwAAQihAEgECEMAIEIYQAIRAgDQCBCGAACEcIAEIgQBoBAhDAABCKEASAQIQwAgQhhAAhECANAIEIYAAIRwgAQiBAGgECEMAAEIoQBIBAhDACBCGEACEQIA0AgQhgAAhHCABCIEAaAQIQwAAQihAEgECEMAIEIYQAIRAgDQCBCGAACEcIAEIgQBoBAhDAABEq1Y5BMJtOOYUKcyecGoPXaEsJbtmxpxzAhzuRzA9B6LEcAQCBCGAACEcIAEIgQBoBAhDAABCKEASAQIQwAgQhhAAhECANAIEIYAAIRwgAQiBAGgECEMAAEIoQBIBAhDACBCGEACEQIA0AgQhgAAhHCABCIEAaAQIQwAAQihAEgECEMAIEIYQAIRAgDQCBCGAACEcIAEIgQBoBAhDAABCKEASAQIQwAgQhhAAhECANAIEIYAAIRwgAQiBAGgECEMAAEIoQBIBAhDACBCGEACJSKLqBU95uv6tzD4yXPZySprG0px5BWN7s0AGiJjgnhTCazoO348XlJUn9/I6G6uuqxAKATdUwIb9myJboEAGg71oQBIBAhDACBCGEACEQIA0AgQhgAAhHCABCIEAaAQIQwAAQihAEgECEMAIEIYQAIRAgDQCBCGAACEcIAEIgQBoBAhDAABCKEASAQIQwAgQhhAAhECANAIHP3pXc2e1nS0WWOdbGkV5a5bxRqbr3TrV6JmtvlTKr5Mne/pNoODYXwSpjZPndf35bBmoSaW+90q1ei5nY5W2pmOQIAAhHCABConSF8TxvHahZqbr3TrV6JmtvlrKi5bWvCAICFWI4AgECEMAAEankIm9mAmX3fzHJm9qlWj9cMZjZtZgfN7Ckz2xddTzVmdq+ZvWRmz5S0XWRme8zsueT/3sgaK9WoeYeZHU/m+ikz2xRZYyUzu9TM9prZITP7npkNJ+0dOdd16u3YeTazVWb2HTM7kNR8W9LekXMs1a254Xlu6ZqwmXVLelbSRknPS3pC0ofdfaplgzaBmU1LWu/uHftGcTP7DUknJI25+xVJ205Jr7r7p5MfeL3u/ueRdZaqUfMOSSfc/a8ja6vFzNZIWuPuT5rZBZL2S/pdSR9RB851nXo/pA6dZzMzSee5+wkz65H0bUnDkn5PHTjHUt2aB9TgPLf6SvgqSTl3P+LuP5H0RUnvb/GYZwV3/6akVyua3y9pd/J4t/LffB2jRs0dzd1fcPcnk8evSzokqV8dOtd16u1YnnciedqT/HN16BxLdWtuWKtDuF/SsZLnz6vDvyASLunrZrbfzD4aXUwDVrv7C1L+m1HSTwfXs1SfMLOnk+WKjvmVs5KZpSX9oqTHdRrMdUW9UgfPs5l1m9lTkl6StMfdO36Oa9QsNTjPrQ5hq9J2Orwn7mp3/yVJg5L+NPk1Gq3xeUnvlnSlpBck/U1sOdWZ2fmSHpT0Z+7+4+h6FlOl3o6eZ3c/6e5XSlor6SozuyK6psXUqLnheW51CD8v6dKS52sl/aDFY66Yu/8g+f8lSQ8pv6xyOngxWRMsrA2+FFzPotz9xeSL+ZSkv1cHznWy5vegpAfc/V+S5o6d62r1ng7zLEnu/kNJjym/ttqxc1yqtOblzHOrQ/gJST9jZuvM7G2S/kDS11o85oqY2XnJCxoys/Mk/ZakZ+rv1TG+JmkoeTwk6V8Da1mSwjdZ4gPqsLlOXoD5B0mH3P1vSzZ15FzXqreT59nMLjGzC5PH50q6VtJhdegcS7VrXs48t/wTc8lbNP5OUreke939jpYOuEJm9i7lr34lKSXpC51Ys5n9k6QNyt8670VJ2yV9VdKXJb1T0v9I+qC7d8wLYTVq3qD8r24uaVrSxwrrgJ3AzN4n6VuSDko6lTT/hfLrrB0313Xq/bA6dJ7N7BeUf+GtW/kLwy+7+1+ZWZ86cI6lujX/oxqcZz62DACB+MQcAAQihAEgECEMAIEIYQAIRAgDQCBCGAACEcJoCjO70Mz+pM72/2zCGB8xs88mjz9uZpvr9N1gZr+20jGBViOE0SwXSloQwsntTOXuTQ1Ed7/b3cfqdNkgiRBGxyOE0SyflvTu5EbWTyQ3Fv+C8p/ckpmdSP7fYGbfNLOHzGzKzO42s5pfh2b2R2b2rJl9Q9LVJe07zOyTyeObkmM9bWZfTO4e9nFJNyf1/LqZ/Y6ZPW5m3zWzR81sdclx7jWzx8zsiJndVDLG5uSYB5JPQhU+rvpgco5PmNnVAlYgFV0AzhifknSFu19pZhskPZI8/+8qfa+S9B5JRyVNKn/z7q9Udko+h3+bpPdK+pGkvZK+W2Psde7+lpld6O4/NLO7VXJz7eSWgr/i7m5mfyzpFklbk/0vl3SNpAskfd/MPi/pZyX9pfJ31HvFzC5K+u6S9Bl3/7aZvVPSv0n6uaVPE1COEEarfKdGABe2HZGK95N4n6qEsKRflvSYu7+c9P2S8uFY6WlJD5jZV5W/f0Y1ayV9KQn2t0kqre0Rd39L0ltm9pKk1ZJ+U9JXCn9dpeSeBddKek/+PjmSpHeY2QXJDdSBhrEcgVZ5o862yhuW1LuByVJubnKdpM8pf8W838yqXVyMSvqsu/+8pI9JWlWy7a2SxyeVvzixGmN3SfpVd78y+ddPAGMlCGE0y+vK/zq/FFcltzftkvT7yv99rmoel7TBzPqSe+R+sLJDcoxL3X2v8ksMF0o6v0o9PyXpePJ4SIv7d0kfSu7kpZLliK9L+kTJ+Fcu4VhATYQwmsLdZyT9h+X/kvKdi3T/L+VfyHtG+WWBh6p1Sm4BuCPp/6ikJ6t065Z0v5kdVH69+DPJTbYflvSBwgtzyXH+2cy+JWnRP+Dq7t+TdIekb5jZAUmFe/PeJGl98oLdlPIvAALLxq0s0VbJi3afdPffjq4F6ARcCQNAIK6E0RHM7HFJ51Q0/6G7H4yoB2gXQhgAArEcAQCBCGEACEQIA0AgQhgAAv0/7Z6WuYTw7WkAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "### YOUR CODE HERE ###\n", "\n", "tdbp = sns.boxplot(df0['trip_distance'])\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fabp = sns.boxplot(df0['fare_amount'])\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "tabp = sns.boxplot(df0['total_amount'])\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "dbp = sns.boxplot(df0['duration'])\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "id": "pqcGiHLa4TvP" }, "source": [ "**Questions:** \n", "1. Which variable(s) contains outliers? \n", "\n", "2. Are the values in the `trip_distance` column unbelievable?\n", "\n", "3. What about the lower end? Do distances, fares, and durations of 0 (or negative values) make sense?" ] }, { "cell_type": "markdown", "metadata": { "id": "FetTHatPoR6n" }, "source": [ "==> ENTER YOUR RESPONSE HERE\n", "All of the selected variables contain outliers. The trip distances values doesn't contain any unbelievable values, except for the minimum trip distaces value, 0, yet the max although an outlier is within reason, 33.\n", "All of the selected variables contain either a 0 or a negative value which doesn't make any sense." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Task 2e. Imputations" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### `trip_distance` outliers\n", "\n", "You know from the summary statistics that there are trip distances of 0. Are these reflective of erroneous data, or are they very short trips that get rounded down?\n", "\n", "To check, sort the column values, eliminate duplicates, and inspect the least 10 values. Are they rounded values or precise values?" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0VendorIDtpep_pickup_datetimetpep_dropoff_datetimepassenger_counttrip_distanceRatecodeIDstore_and_fwd_flagPULocationIDDOLocationIDpayment_typefare_amountextramta_taxtip_amounttolls_amountimprovement_surchargetotal_amountduration
220266364292312017-07-27 07:44:242017-07-27 07:44:2410.001N41264210.500.00.50.000.000.311.300.000000
79510113503012017-11-30 07:11:342017-11-30 07:11:3410.001N24626428.000.00.50.000.000.38.800.000000
69082416204522017-03-26 02:07:082017-03-26 02:07:1210.005N6161118.000.00.02.000.000.320.300.066667
135611450436512017-02-23 16:06:312017-02-23 16:06:5420.005N175175332.000.00.00.000.000.332.300.383333
122389554492312017-11-11 09:28:132017-11-11 09:28:2720.001N14514522.500.00.50.000.000.33.300.233333
............................................................
299405244622017-11-06 20:30:502017-11-07 00:00:00130.831N13223180.000.50.518.5611.520.3111.38209.166667
102917631933022017-09-11 11:41:042017-09-11 12:18:58131.954N1382652131.000.00.50.000.000.3131.8037.900000
60644989402322017-06-13 12:30:222017-06-13 13:37:51132.723N13811107.000.00.055.5016.260.3179.0667.483333
138614052366822017-05-19 08:20:212017-05-19 09:20:30133.925N2292651200.010.00.551.645.760.3258.2160.150000
92805181071422017-06-18 23:33:252017-06-19 00:12:38233.965N1322652150.000.00.00.000.000.3150.3039.216667
\n", "

22699 rows × 19 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 VendorID tpep_pickup_datetime tpep_dropoff_datetime \\\n", "22026 63642923 1 2017-07-27 07:44:24 2017-07-27 07:44:24 \n", "795 101135030 1 2017-11-30 07:11:34 2017-11-30 07:11:34 \n", "6908 24162045 2 2017-03-26 02:07:08 2017-03-26 02:07:12 \n", "13561 14504365 1 2017-02-23 16:06:31 2017-02-23 16:06:54 \n", "12238 95544923 1 2017-11-11 09:28:13 2017-11-11 09:28:27 \n", "... ... ... ... ... \n", "29 94052446 2 2017-11-06 20:30:50 2017-11-07 00:00:00 \n", "10291 76319330 2 2017-09-11 11:41:04 2017-09-11 12:18:58 \n", "6064 49894023 2 2017-06-13 12:30:22 2017-06-13 13:37:51 \n", "13861 40523668 2 2017-05-19 08:20:21 2017-05-19 09:20:30 \n", "9280 51810714 2 2017-06-18 23:33:25 2017-06-19 00:12:38 \n", "\n", " passenger_count trip_distance RatecodeID store_and_fwd_flag \\\n", "22026 1 0.00 1 N \n", "795 1 0.00 1 N \n", "6908 1 0.00 5 N \n", "13561 2 0.00 5 N \n", "12238 2 0.00 1 N \n", "... ... ... ... ... \n", "29 1 30.83 1 N \n", "10291 1 31.95 4 N \n", "6064 1 32.72 3 N \n", "13861 1 33.92 5 N \n", "9280 2 33.96 5 N \n", "\n", " PULocationID DOLocationID payment_type fare_amount extra mta_tax \\\n", "22026 41 264 2 10.50 0.0 0.5 \n", "795 246 264 2 8.00 0.0 0.5 \n", "6908 61 61 1 18.00 0.0 0.0 \n", "13561 175 175 3 32.00 0.0 0.0 \n", "12238 145 145 2 2.50 0.0 0.5 \n", "... ... ... ... ... ... ... \n", "29 132 23 1 80.00 0.5 0.5 \n", "10291 138 265 2 131.00 0.0 0.5 \n", "6064 138 1 1 107.00 0.0 0.0 \n", "13861 229 265 1 200.01 0.0 0.5 \n", "9280 132 265 2 150.00 0.0 0.0 \n", "\n", " tip_amount tolls_amount improvement_surcharge total_amount \\\n", "22026 0.00 0.00 0.3 11.30 \n", "795 0.00 0.00 0.3 8.80 \n", "6908 2.00 0.00 0.3 20.30 \n", "13561 0.00 0.00 0.3 32.30 \n", "12238 0.00 0.00 0.3 3.30 \n", "... ... ... ... ... \n", "29 18.56 11.52 0.3 111.38 \n", "10291 0.00 0.00 0.3 131.80 \n", "6064 55.50 16.26 0.3 179.06 \n", "13861 51.64 5.76 0.3 258.21 \n", "9280 0.00 0.00 0.3 150.30 \n", "\n", " duration \n", "22026 0.000000 \n", "795 0.000000 \n", "6908 0.066667 \n", "13561 0.383333 \n", "12238 0.233333 \n", "... ... \n", "29 209.166667 \n", "10291 37.900000 \n", "6064 67.483333 \n", "13861 60.150000 \n", "9280 39.216667 \n", "\n", "[22699 rows x 19 columns]" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Are trip distances of 0 bad data or very short trips rounded down?\n", "### YOUR CODE HERE ###\n", "td = df0.sort_values(by='trip_distance')\n", "td.drop_duplicates()\n", "td" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The distances are captured with a high degree of precision. However, it might be possible for trips to have distances of zero if a passenger summoned a taxi and then changed their mind. Besides, are there enough zero values in the data to pose a problem?\n", "\n", "Calculate the count of rides where the `trip_distance` is zero." ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "148" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "### YOUR CODE HERE ###\n", "df0['trip_distance'][df0['trip_distance']==0].count()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### `fare_amount` outliers" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0VendorIDtpep_pickup_datetimetpep_dropoff_datetimepassenger_counttrip_distanceRatecodeIDstore_and_fwd_flagPULocationIDDOLocationIDpayment_typefare_amountextramta_taxtip_amounttolls_amountimprovement_surchargetotal_amountduration
31410545428722017-12-13 02:02:392017-12-13 02:03:0860.121N1611613-2.5-0.5-0.50.00.0-0.3-3.80.483333
16465733718322017-07-05 11:02:232017-07-05 11:03:0010.041N79793-2.50.0-0.50.00.0-0.3-3.30.616667
440210801695422017-12-20 16:06:532017-12-20 16:47:5017.061N26316920.00.00.00.00.00.00.040.950000
44239732990522017-11-16 20:13:302017-11-16 20:14:5020.061N2372374-3.0-0.5-0.50.00.0-0.3-4.31.333333
54482845998322017-04-06 12:50:262017-04-06 12:52:3910.251N90683-3.50.0-0.50.00.0-0.3-4.32.216667
57224967036422017-06-12 12:08:552017-06-12 12:08:5710.001N26419310.00.00.00.00.00.00.00.033333
575883394822017-01-03 20:15:232017-01-03 20:15:3910.021N1701703-2.5-0.5-0.50.00.0-0.3-3.80.266667
82049118794722017-10-28 20:39:362017-10-28 20:41:5910.411N2362373-3.5-0.5-0.50.00.0-0.3-4.82.383333
102815530234722017-06-05 17:34:252017-06-05 17:36:2920.001N2382384-2.5-1.0-0.50.00.0-0.3-4.32.066667
105062600502422017-03-30 03:14:262017-03-30 03:14:2810.001N26419310.00.00.00.00.00.00.00.033333
112045839550122017-07-09 07:20:592017-07-09 07:23:5010.641N50483-4.50.0-0.50.00.0-0.3-5.32.850000
129442905976022017-04-08 00:00:162017-04-08 23:15:5710.175N1381384-120.00.00.00.00.0-0.3-120.31395.683333
1471410927609222017-12-24 22:37:582017-12-24 22:41:0850.401N1641614-4.0-0.5-0.50.00.0-0.3-5.33.166667
176022469014622017-03-24 19:31:132017-03-24 19:34:4910.461N87454-4.0-1.0-0.50.00.0-0.3-5.83.600000
185654385976022017-05-22 15:51:202017-05-22 15:52:2210.101N2301633-3.00.0-0.50.00.0-0.3-3.81.033333
190675871301912017-07-10 14:40:092017-07-10 14:40:5910.105N2611330.00.00.00.00.00.30.30.833333
203177592691522017-09-09 22:59:512017-09-09 23:02:0610.241N1161164-3.5-0.5-0.50.00.0-0.3-4.82.250000
206981466820922017-02-24 00:38:172017-02-24 00:42:0510.701N65254-4.5-0.5-0.50.00.0-0.3-5.83.800000
218423170808312017-04-18 16:55:292017-04-18 18:29:44220.405N26426430.00.00.00.012.50.312.894.250000
225661902289822017-03-07 02:24:472017-03-07 02:24:5010.001N26419310.00.00.00.00.00.00.00.050000
\n", "
" ], "text/plain": [ " Unnamed: 0 VendorID tpep_pickup_datetime tpep_dropoff_datetime \\\n", "314 105454287 2 2017-12-13 02:02:39 2017-12-13 02:03:08 \n", "1646 57337183 2 2017-07-05 11:02:23 2017-07-05 11:03:00 \n", "4402 108016954 2 2017-12-20 16:06:53 2017-12-20 16:47:50 \n", "4423 97329905 2 2017-11-16 20:13:30 2017-11-16 20:14:50 \n", "5448 28459983 2 2017-04-06 12:50:26 2017-04-06 12:52:39 \n", "5722 49670364 2 2017-06-12 12:08:55 2017-06-12 12:08:57 \n", "5758 833948 2 2017-01-03 20:15:23 2017-01-03 20:15:39 \n", "8204 91187947 2 2017-10-28 20:39:36 2017-10-28 20:41:59 \n", "10281 55302347 2 2017-06-05 17:34:25 2017-06-05 17:36:29 \n", "10506 26005024 2 2017-03-30 03:14:26 2017-03-30 03:14:28 \n", "11204 58395501 2 2017-07-09 07:20:59 2017-07-09 07:23:50 \n", "12944 29059760 2 2017-04-08 00:00:16 2017-04-08 23:15:57 \n", "14714 109276092 2 2017-12-24 22:37:58 2017-12-24 22:41:08 \n", "17602 24690146 2 2017-03-24 19:31:13 2017-03-24 19:34:49 \n", "18565 43859760 2 2017-05-22 15:51:20 2017-05-22 15:52:22 \n", "19067 58713019 1 2017-07-10 14:40:09 2017-07-10 14:40:59 \n", "20317 75926915 2 2017-09-09 22:59:51 2017-09-09 23:02:06 \n", "20698 14668209 2 2017-02-24 00:38:17 2017-02-24 00:42:05 \n", "21842 31708083 1 2017-04-18 16:55:29 2017-04-18 18:29:44 \n", "22566 19022898 2 2017-03-07 02:24:47 2017-03-07 02:24:50 \n", "\n", " passenger_count trip_distance RatecodeID store_and_fwd_flag \\\n", "314 6 0.12 1 N \n", "1646 1 0.04 1 N \n", "4402 1 7.06 1 N \n", "4423 2 0.06 1 N \n", "5448 1 0.25 1 N \n", "5722 1 0.00 1 N \n", "5758 1 0.02 1 N \n", "8204 1 0.41 1 N \n", "10281 2 0.00 1 N \n", "10506 1 0.00 1 N \n", "11204 1 0.64 1 N \n", "12944 1 0.17 5 N \n", "14714 5 0.40 1 N \n", "17602 1 0.46 1 N \n", "18565 1 0.10 1 N \n", "19067 1 0.10 5 N \n", "20317 1 0.24 1 N \n", "20698 1 0.70 1 N \n", "21842 2 20.40 5 N \n", "22566 1 0.00 1 N \n", "\n", " PULocationID DOLocationID payment_type fare_amount extra mta_tax \\\n", "314 161 161 3 -2.5 -0.5 -0.5 \n", "1646 79 79 3 -2.5 0.0 -0.5 \n", "4402 263 169 2 0.0 0.0 0.0 \n", "4423 237 237 4 -3.0 -0.5 -0.5 \n", "5448 90 68 3 -3.5 0.0 -0.5 \n", "5722 264 193 1 0.0 0.0 0.0 \n", "5758 170 170 3 -2.5 -0.5 -0.5 \n", "8204 236 237 3 -3.5 -0.5 -0.5 \n", "10281 238 238 4 -2.5 -1.0 -0.5 \n", "10506 264 193 1 0.0 0.0 0.0 \n", "11204 50 48 3 -4.5 0.0 -0.5 \n", "12944 138 138 4 -120.0 0.0 0.0 \n", "14714 164 161 4 -4.0 -0.5 -0.5 \n", "17602 87 45 4 -4.0 -1.0 -0.5 \n", "18565 230 163 3 -3.0 0.0 -0.5 \n", "19067 261 13 3 0.0 0.0 0.0 \n", "20317 116 116 4 -3.5 -0.5 -0.5 \n", "20698 65 25 4 -4.5 -0.5 -0.5 \n", "21842 264 264 3 0.0 0.0 0.0 \n", "22566 264 193 1 0.0 0.0 0.0 \n", "\n", " tip_amount tolls_amount improvement_surcharge total_amount \\\n", "314 0.0 0.0 -0.3 -3.8 \n", "1646 0.0 0.0 -0.3 -3.3 \n", "4402 0.0 0.0 0.0 0.0 \n", "4423 0.0 0.0 -0.3 -4.3 \n", "5448 0.0 0.0 -0.3 -4.3 \n", "5722 0.0 0.0 0.0 0.0 \n", "5758 0.0 0.0 -0.3 -3.8 \n", "8204 0.0 0.0 -0.3 -4.8 \n", "10281 0.0 0.0 -0.3 -4.3 \n", "10506 0.0 0.0 0.0 0.0 \n", "11204 0.0 0.0 -0.3 -5.3 \n", "12944 0.0 0.0 -0.3 -120.3 \n", "14714 0.0 0.0 -0.3 -5.3 \n", "17602 0.0 0.0 -0.3 -5.8 \n", "18565 0.0 0.0 -0.3 -3.8 \n", "19067 0.0 0.0 0.3 0.3 \n", "20317 0.0 0.0 -0.3 -4.8 \n", "20698 0.0 0.0 -0.3 -5.8 \n", "21842 0.0 12.5 0.3 12.8 \n", "22566 0.0 0.0 0.0 0.0 \n", "\n", " duration \n", "314 0.483333 \n", "1646 0.616667 \n", "4402 40.950000 \n", "4423 1.333333 \n", "5448 2.216667 \n", "5722 0.033333 \n", "5758 0.266667 \n", "8204 2.383333 \n", "10281 2.066667 \n", "10506 0.033333 \n", "11204 2.850000 \n", "12944 1395.683333 \n", "14714 3.166667 \n", "17602 3.600000 \n", "18565 1.033333 \n", "19067 0.833333 \n", "20317 2.250000 \n", "20698 3.800000 \n", "21842 94.250000 \n", "22566 0.050000 " ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "### YOUR CODE HERE ###\n", "df0[df0['fare_amount']<=0]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Question:** What do you notice about the values in the `fare_amount` column?\n", "\n", "Impute values less than $0 with `0`." ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0VendorIDtpep_pickup_datetimetpep_dropoff_datetimepassenger_counttrip_distanceRatecodeIDstore_and_fwd_flagPULocationIDDOLocationIDpayment_typefare_amountextramta_taxtip_amounttolls_amountimprovement_surchargetotal_amountduration
31410545428722017-12-13 02:02:392017-12-13 02:03:0860.121N1611613-2.5-0.5-0.50.00.0-0.3-3.80.483333
16465733718322017-07-05 11:02:232017-07-05 11:03:0010.041N79793-2.50.0-0.50.00.0-0.3-3.30.616667
440210801695422017-12-20 16:06:532017-12-20 16:47:5017.061N26316920.00.00.00.00.00.00.040.950000
44239732990522017-11-16 20:13:302017-11-16 20:14:5020.061N2372374-3.0-0.5-0.50.00.0-0.3-4.31.333333
54482845998322017-04-06 12:50:262017-04-06 12:52:3910.251N90683-3.50.0-0.50.00.0-0.3-4.32.216667
57224967036422017-06-12 12:08:552017-06-12 12:08:5710.001N26419310.00.00.00.00.00.00.00.033333
575883394822017-01-03 20:15:232017-01-03 20:15:3910.021N1701703-2.5-0.5-0.50.00.0-0.3-3.80.266667
82049118794722017-10-28 20:39:362017-10-28 20:41:5910.411N2362373-3.5-0.5-0.50.00.0-0.3-4.82.383333
102815530234722017-06-05 17:34:252017-06-05 17:36:2920.001N2382384-2.5-1.0-0.50.00.0-0.3-4.32.066667
105062600502422017-03-30 03:14:262017-03-30 03:14:2810.001N26419310.00.00.00.00.00.00.00.033333
112045839550122017-07-09 07:20:592017-07-09 07:23:5010.641N50483-4.50.0-0.50.00.0-0.3-5.32.850000
129442905976022017-04-08 00:00:162017-04-08 23:15:5710.175N1381384-120.00.00.00.00.0-0.3-120.31395.683333
1471410927609222017-12-24 22:37:582017-12-24 22:41:0850.401N1641614-4.0-0.5-0.50.00.0-0.3-5.33.166667
176022469014622017-03-24 19:31:132017-03-24 19:34:4910.461N87454-4.0-1.0-0.50.00.0-0.3-5.83.600000
185654385976022017-05-22 15:51:202017-05-22 15:52:2210.101N2301633-3.00.0-0.50.00.0-0.3-3.81.033333
190675871301912017-07-10 14:40:092017-07-10 14:40:5910.105N2611330.00.00.00.00.00.30.30.833333
203177592691522017-09-09 22:59:512017-09-09 23:02:0610.241N1161164-3.5-0.5-0.50.00.0-0.3-4.82.250000
206981466820922017-02-24 00:38:172017-02-24 00:42:0510.701N65254-4.5-0.5-0.50.00.0-0.3-5.83.800000
218423170808312017-04-18 16:55:292017-04-18 18:29:44220.405N26426430.00.00.00.012.50.312.894.250000
225661902289822017-03-07 02:24:472017-03-07 02:24:5010.001N26419310.00.00.00.00.00.00.00.050000
\n", "
" ], "text/plain": [ " Unnamed: 0 VendorID tpep_pickup_datetime tpep_dropoff_datetime \\\n", "314 105454287 2 2017-12-13 02:02:39 2017-12-13 02:03:08 \n", "1646 57337183 2 2017-07-05 11:02:23 2017-07-05 11:03:00 \n", "4402 108016954 2 2017-12-20 16:06:53 2017-12-20 16:47:50 \n", "4423 97329905 2 2017-11-16 20:13:30 2017-11-16 20:14:50 \n", "5448 28459983 2 2017-04-06 12:50:26 2017-04-06 12:52:39 \n", "5722 49670364 2 2017-06-12 12:08:55 2017-06-12 12:08:57 \n", "5758 833948 2 2017-01-03 20:15:23 2017-01-03 20:15:39 \n", "8204 91187947 2 2017-10-28 20:39:36 2017-10-28 20:41:59 \n", "10281 55302347 2 2017-06-05 17:34:25 2017-06-05 17:36:29 \n", "10506 26005024 2 2017-03-30 03:14:26 2017-03-30 03:14:28 \n", "11204 58395501 2 2017-07-09 07:20:59 2017-07-09 07:23:50 \n", "12944 29059760 2 2017-04-08 00:00:16 2017-04-08 23:15:57 \n", "14714 109276092 2 2017-12-24 22:37:58 2017-12-24 22:41:08 \n", "17602 24690146 2 2017-03-24 19:31:13 2017-03-24 19:34:49 \n", "18565 43859760 2 2017-05-22 15:51:20 2017-05-22 15:52:22 \n", "19067 58713019 1 2017-07-10 14:40:09 2017-07-10 14:40:59 \n", "20317 75926915 2 2017-09-09 22:59:51 2017-09-09 23:02:06 \n", "20698 14668209 2 2017-02-24 00:38:17 2017-02-24 00:42:05 \n", "21842 31708083 1 2017-04-18 16:55:29 2017-04-18 18:29:44 \n", "22566 19022898 2 2017-03-07 02:24:47 2017-03-07 02:24:50 \n", "\n", " passenger_count trip_distance RatecodeID store_and_fwd_flag \\\n", "314 6 0.12 1 N \n", "1646 1 0.04 1 N \n", "4402 1 7.06 1 N \n", "4423 2 0.06 1 N \n", "5448 1 0.25 1 N \n", "5722 1 0.00 1 N \n", "5758 1 0.02 1 N \n", "8204 1 0.41 1 N \n", "10281 2 0.00 1 N \n", "10506 1 0.00 1 N \n", "11204 1 0.64 1 N \n", "12944 1 0.17 5 N \n", "14714 5 0.40 1 N \n", "17602 1 0.46 1 N \n", "18565 1 0.10 1 N \n", "19067 1 0.10 5 N \n", "20317 1 0.24 1 N \n", "20698 1 0.70 1 N \n", "21842 2 20.40 5 N \n", "22566 1 0.00 1 N \n", "\n", " PULocationID DOLocationID payment_type fare_amount extra mta_tax \\\n", "314 161 161 3 -2.5 -0.5 -0.5 \n", "1646 79 79 3 -2.5 0.0 -0.5 \n", "4402 263 169 2 0.0 0.0 0.0 \n", "4423 237 237 4 -3.0 -0.5 -0.5 \n", "5448 90 68 3 -3.5 0.0 -0.5 \n", "5722 264 193 1 0.0 0.0 0.0 \n", "5758 170 170 3 -2.5 -0.5 -0.5 \n", "8204 236 237 3 -3.5 -0.5 -0.5 \n", "10281 238 238 4 -2.5 -1.0 -0.5 \n", "10506 264 193 1 0.0 0.0 0.0 \n", "11204 50 48 3 -4.5 0.0 -0.5 \n", "12944 138 138 4 -120.0 0.0 0.0 \n", "14714 164 161 4 -4.0 -0.5 -0.5 \n", "17602 87 45 4 -4.0 -1.0 -0.5 \n", "18565 230 163 3 -3.0 0.0 -0.5 \n", "19067 261 13 3 0.0 0.0 0.0 \n", "20317 116 116 4 -3.5 -0.5 -0.5 \n", "20698 65 25 4 -4.5 -0.5 -0.5 \n", "21842 264 264 3 0.0 0.0 0.0 \n", "22566 264 193 1 0.0 0.0 0.0 \n", "\n", " tip_amount tolls_amount improvement_surcharge total_amount \\\n", "314 0.0 0.0 -0.3 -3.8 \n", "1646 0.0 0.0 -0.3 -3.3 \n", "4402 0.0 0.0 0.0 0.0 \n", "4423 0.0 0.0 -0.3 -4.3 \n", "5448 0.0 0.0 -0.3 -4.3 \n", "5722 0.0 0.0 0.0 0.0 \n", "5758 0.0 0.0 -0.3 -3.8 \n", "8204 0.0 0.0 -0.3 -4.8 \n", "10281 0.0 0.0 -0.3 -4.3 \n", "10506 0.0 0.0 0.0 0.0 \n", "11204 0.0 0.0 -0.3 -5.3 \n", "12944 0.0 0.0 -0.3 -120.3 \n", "14714 0.0 0.0 -0.3 -5.3 \n", "17602 0.0 0.0 -0.3 -5.8 \n", "18565 0.0 0.0 -0.3 -3.8 \n", "19067 0.0 0.0 0.3 0.3 \n", "20317 0.0 0.0 -0.3 -4.8 \n", "20698 0.0 0.0 -0.3 -5.8 \n", "21842 0.0 12.5 0.3 12.8 \n", "22566 0.0 0.0 0.0 0.0 \n", "\n", " duration \n", "314 0.483333 \n", "1646 0.616667 \n", "4402 40.950000 \n", "4423 1.333333 \n", "5448 2.216667 \n", "5722 0.033333 \n", "5758 0.266667 \n", "8204 2.383333 \n", "10281 2.066667 \n", "10506 0.033333 \n", "11204 2.850000 \n", "12944 1395.683333 \n", "14714 3.166667 \n", "17602 3.600000 \n", "18565 1.033333 \n", "19067 0.833333 \n", "20317 2.250000 \n", "20698 3.800000 \n", "21842 94.250000 \n", "22566 0.050000 " ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Impute values less than $0 with 0\n", "### YOUR CODE HERE ###\n", "df0[(df0['fare_amount']<=0)]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now impute the maximum value as `Q3 + (6 * IQR)`." ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [], "source": [ "### YOUR CODE HERE ###\n", "'''\n", "Impute upper-limit values in specified columns based on their interquartile range.\n", "\n", "Arguments:\n", " column_list: A list of columns to iterate over\n", " iqr_factor: A number representing x in the formula:\n", " Q3 + (x * IQR). Used to determine maximum threshold,\n", " beyond which a point is considered an outlier.\n", "\n", "The IQR is computed for each column in column_list and values exceeding\n", "the upper threshold for each column are imputed with the upper threshold value.\n", "'''\n", "### YOUR CODE HERE ###\n", "# Reassign minimum to zero\n", "### YOUR CODE HERE ###\n", "df0['fare_amount'][(df0['fare_amount']<=0)] = 0\n", "\n", "# Calculate upper threshold\n", "### YOUR CODE HERE ###\n", "q1 = df0['fare_amount'].quantile(0.25)\n", "q3 = df0['fare_amount'].quantile(0.75)\n", "iqr = q3 - q1\n", "th = q3 + (6*iqr)\n", "# Reassign values > threshold to threshold\n", "### YOUR CODE HERE ###\n", "df0['fare_amount'][(df0['fare_amount']>th)] = th" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fabp = sns.boxplot(df0['fare_amount'])\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### `duration` outliers\n" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "count 22699.000000\n", "mean 17.013777\n", "std 61.996482\n", "min -16.983333\n", "25% 6.650000\n", "50% 11.183333\n", "75% 18.383333\n", "max 1439.550000\n", "Name: duration, dtype: float64" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Call .describe() for duration outliers\n", "### YOUR CODE HERE ###\n", "df0['duration'].describe()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The `duration` column has problematic values at both the lower and upper extremities.\n", "\n", "* **Low values:** There should be no values that represent negative time. Impute all negative durations with `0`.\n", "\n", "* **High values:** Impute high values the same way you imputed the high-end outliers for fares: `Q3 + (6 * IQR)`." ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [], "source": [ "# Impute a 0 for any negative values\n", "### YOUR CODE HERE ###### YOUR CODE HERE ###\n", "# Reassign minimum to zero\n", "### YOUR CODE HERE ###\n", "df0['duration'][(df0['duration']<=0)] = 0" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "# Impute the high outliers\n", "### YOUR CODE HERE ###\n", "# Calculate upper threshold\n", "### YOUR CODE HERE ###\n", "q1 = df0['duration'].quantile(0.25)\n", "q3 = df0['duration'].quantile(0.75)\n", "iqr = q3 - q1\n", "th = q3 + (6*iqr)\n", "# Reassign values > threshold to threshold\n", "### YOUR CODE HERE ###\n", "df0['duration'][(df0['duration']>th)] = th" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "sns.boxplot(df0['duration'])\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "sns.boxplot(df0['fare_amount'])\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Task 3a. Feature engineering" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Create `mean_distance` column\n", "\n", "When deployed, the model will not know the duration of a trip until after the trip occurs, so you cannot train a model that uses this feature. However, you can use the statistics of trips you *do* know to generalize about ones you do not know.\n", "\n", "In this step, create a column called `mean_distance` that captures the mean distance for each group of trips that share pickup and dropoff points.\n", "\n", "For example, if your data were:\n", "\n", "|Trip|Start|End|Distance|\n", "|--: |:---:|:-:| |\n", "| 1 | A | B | 1 |\n", "| 2 | C | D | 2 |\n", "| 3 | A | B |1.5 |\n", "| 4 | D | C | 3 |\n", "\n", "The results should be:\n", "```\n", "A -> B: 1.25 miles\n", "C -> D: 2 miles\n", "D -> C: 3 miles\n", "```\n", "\n", "Notice that C -> D is not the same as D -> C. All trips that share a unique pair of start and end points get grouped and averaged.\n", "\n", "Then, a new column `mean_distance` will be added where the value at each row is the average for all trips with those pickup and dropoff locations:\n", "\n", "|Trip|Start|End|Distance|mean_distance|\n", "|--: |:---:|:-:| :-- |:-- |\n", "| 1 | A | B | 1 | 1.25 |\n", "| 2 | C | D | 2 | 2 |\n", "| 3 | A | B |1.5 | 1.25 |\n", "| 4 | D | C | 3 | 3 |\n", "\n", "\n", "Begin by creating a helper column called `pickup_dropoff`, which contains the unique combination of pickup and dropoff location IDs for each row.\n", "\n", "One way to do this is to convert the pickup and dropoff location IDs to strings and join them, separated by a space. The space is to ensure that, for example, a trip with pickup/dropoff points of 12 & 151 gets encoded differently than a trip with points 121 & 51.\n", "\n", "So, the new column would look like this:\n", "\n", "|Trip|Start|End|pickup_dropoff|\n", "|--: |:---:|:-:| :-- |\n", "| 1 | A | B | 'A B' |\n", "| 2 | C | D | 'C D' |\n", "| 3 | A | B | 'A B' |\n", "| 4 | D | C | 'D C' |\n" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 100 231\n", "1 186 43\n", "2 262 236\n", "3 188 97\n", "4 4 112\n", " ... \n", "22694 48 186\n", "22695 132 164\n", "22696 107 234\n", "22697 68 144\n", "22698 239 236\n", "Name: pickup_dropoff, Length: 22699, dtype: object" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Create `pickup_dropoff` column\n", "### YOUR CODE HERE ###\n", "df0['pickup_dropoff'] = df0['PULocationID'].astype(str) + ' ' + df0['DOLocationID'].astype(str)\n", "df0['pickup_dropoff']" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now, use a `groupby()` statement to group each row by the new `pickup_dropoff` column, compute the mean, and capture the values only in the `trip_distance` column. Assign the results to a variable named `grouped`." ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "pickup_dropoff\n", "1 1 2.433333\n", "10 148 15.700000\n", "100 1 16.890000\n", "100 100 0.253333\n", "100 107 1.180000\n", " ... \n", "97 65 0.500000\n", "97 66 1.400000\n", "97 80 3.840000\n", "97 90 4.420000\n", "97 97 1.006667\n", "Name: trip_distance, Length: 4172, dtype: float64" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "### YOUR CODE HERE ###\n", "grouped = df0.groupby('pickup_dropoff')['trip_distance'].mean()\n", "grouped" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "`grouped` is an object of the `DataFrame` class.\n", "\n", "1. Convert it to a dictionary using the [`to_dict()`](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.to_dict.html) method. Assign the results to a variable called `grouped_dict`. This will result in a dictionary with a key of `trip_distance` whose values are another dictionary. The inner dictionary's keys are pickup/dropoff points and its values are mean distances. This is the information you want.\n", "\n", "```\n", "Example:\n", "grouped_dict = {'trip_distance': {'A B': 1.25, 'C D': 2, 'D C': 3}\n", "```\n", "\n", "2. Reassign the `grouped_dict` dictionary so it contains only the inner dictionary. In other words, get rid of `trip_distance` as a key, so:\n", "\n", "```\n", "Example:\n", "grouped_dict = {'A B': 1.25, 'C D': 2, 'D C': 3}\n", " ```" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'1 1': 2.433333333333333,\n", " '10 148': 15.7,\n", " '100 1': 16.89,\n", " '100 100': 0.25333333333333335,\n", " '100 107': 1.18,\n", " '100 113': 2.024,\n", " '100 114': 1.94,\n", " '100 12': 4.55,\n", " '100 125': 2.84,\n", " '100 13': 4.201666666666667,\n", " '100 132': 17.2175,\n", " '100 137': 1.299,\n", " '100 138': 10.432857142857143,\n", " '100 140': 2.746,\n", " '100 141': 2.11,\n", " '100 142': 1.6958333333333335,\n", " '100 143': 1.5825,\n", " '100 144': 3.0066666666666664,\n", " '100 148': 4.1066666666666665,\n", " '100 151': 3.668,\n", " '100 152': 4.9,\n", " '100 158': 1.938,\n", " '100 161': 0.9813888888888889,\n", " '100 162': 1.2163636363636363,\n", " '100 163': 1.2656,\n", " '100 164': 0.841,\n", " '100 166': 5.199999999999999,\n", " '100 170': 0.8548,\n", " '100 177': 12.0,\n", " '100 181': 9.34,\n", " '100 186': 0.6404761904761904,\n", " '100 193': 4.39,\n", " '100 198': 9.01,\n", " '100 202': 5.3,\n", " '100 209': 4.43,\n", " '100 211': 2.48,\n", " '100 224': 1.9500000000000002,\n", " '100 225': 7.5,\n", " '100 229': 1.7850000000000001,\n", " '100 230': 0.72975,\n", " '100 231': 3.5216666666666665,\n", " '100 232': 3.8449999999999998,\n", " '100 233': 1.2458333333333333,\n", " '100 234': 1.2545454545454546,\n", " '100 236': 3.3375,\n", " '100 237': 2.5566666666666666,\n", " '100 238': 3.3560000000000003,\n", " '100 239': 2.327142857142857,\n", " '100 243': 8.77,\n", " '100 244': 7.9,\n", " '100 246': 1.1746666666666667,\n", " '100 249': 1.8066666666666666,\n", " '100 25': 7.36,\n", " '100 255': 6.35,\n", " '100 256': 5.859999999999999,\n", " '100 261': 3.8075,\n", " '100 262': 3.8200000000000003,\n", " '100 263': 3.4,\n", " '100 39': 22.6,\n", " '100 4': 2.6999999999999997,\n", " '100 40': 7.23,\n", " '100 41': 4.6,\n", " '100 42': 6.779999999999999,\n", " '100 43': 2.033333333333333,\n", " '100 45': 3.63,\n", " '100 48': 0.8522727272727273,\n", " '100 49': 7.35,\n", " '100 50': 1.1800000000000002,\n", " '100 66': 4.7,\n", " '100 68': 0.9942857142857143,\n", " '100 7': 4.9,\n", " '100 74': 4.53,\n", " '100 75': 4.03,\n", " '100 79': 2.608571428571428,\n", " '100 87': 5.03,\n", " '100 88': 5.495,\n", " '100 90': 1.1228571428571428,\n", " '100 95': 9.0,\n", " '106 106': 0.02,\n", " '106 181': 1.1,\n", " '106 228': 1.24,\n", " '106 231': 3.8,\n", " '106 40': 0.8,\n", " '107 1': 15.55,\n", " '107 100': 1.436,\n", " '107 107': 0.48814814814814816,\n", " '107 113': 0.8969230769230769,\n", " '107 114': 1.207142857142857,\n", " '107 125': 1.8,\n", " '107 127': 11.57,\n", " '107 13': 3.8733333333333335,\n", " '107 130': 12.43,\n", " '107 132': 16.755,\n", " '107 137': 0.6828947368421052,\n", " '107 138': 10.385,\n", " '107 140': 2.801428571428571,\n", " '107 141': 2.981666666666667,\n", " '107 142': 3.228,\n", " '107 143': 4.3,\n", " '107 144': 1.61625,\n", " '107 145': 3.5300000000000002,\n", " '107 146': 4.3,\n", " '107 147': 8.11,\n", " '107 148': 1.7266666666666666,\n", " '107 152': 6.62,\n", " '107 158': 1.7777777777777777,\n", " '107 161': 1.7091666666666667,\n", " '107 162': 1.5677272727272729,\n", " '107 163': 2.4775,\n", " '107 164': 0.748,\n", " '107 170': 1.0014285714285716,\n", " '107 186': 1.4310344827586208,\n", " '107 196': 7.890000000000001,\n", " '107 202': 5.86,\n", " '107 209': 3.496,\n", " '107 21': 11.5,\n", " '107 211': 1.7342857142857144,\n", " '107 223': 5.7,\n", " '107 224': 0.8533333333333334,\n", " '107 229': 1.911111111111111,\n", " '107 23': 17.72,\n", " '107 230': 2.1075,\n", " '107 231': 3.278,\n", " '107 232': 2.1418181818181816,\n", " '107 233': 1.3471428571428572,\n", " '107 234': 0.6842424242424242,\n", " '107 236': 3.165,\n", " '107 237': 2.2516666666666665,\n", " '107 238': 4.986666666666667,\n", " '107 244': 8.84,\n", " '107 246': 1.577142857142857,\n", " '107 249': 1.3976470588235295,\n", " '107 25': 4.5,\n", " '107 256': 3.8666666666666667,\n", " '107 257': 9.0,\n", " '107 26': 10.33,\n", " '107 261': 2.19,\n", " '107 262': 3.5725,\n", " '107 263': 3.66875,\n", " '107 265': 4.5,\n", " '107 36': 5.6,\n", " '107 37': 4.9,\n", " '107 4': 1.16,\n", " '107 41': 5.975,\n", " '107 42': 7.0,\n", " '107 43': 2.8,\n", " '107 45': 2.1375,\n", " '107 48': 2.5460000000000003,\n", " '107 49': 5.0,\n", " '107 66': 3.58,\n", " '107 68': 1.3646666666666667,\n", " '107 7': 5.95,\n", " '107 74': 5.085,\n", " '107 75': 4.930000000000001,\n", " '107 79': 0.9903225806451613,\n", " '107 80': 4.7,\n", " '107 82': 7.32,\n", " '107 87': 3.5,\n", " '107 88': 3.501666666666667,\n", " '107 89': 6.97,\n", " '107 90': 0.9226666666666666,\n", " '112 112': 0.7,\n", " '112 223': 4.05,\n", " '112 263': 8.0,\n", " '112 49': 3.1,\n", " '112 66': 4.57,\n", " '112 80': 0.41500000000000004,\n", " '113 100': 1.9775,\n", " '113 106': 4.69,\n", " '113 107': 1.0516666666666667,\n", " '113 112': 4.5,\n", " '113 113': 0.8346153846153845,\n", " '113 114': 0.7644444444444445,\n", " '113 116': 8.55,\n", " '113 125': 1.2175,\n", " '113 13': 2.2085714285714286,\n", " '113 137': 1.3325,\n", " '113 138': 10.4,\n", " '113 14': 15.62,\n", " '113 140': 3.3,\n", " '113 141': 3.6333333333333333,\n", " '113 142': 3.6,\n", " '113 143': 4.39,\n", " '113 144': 1.1291666666666667,\n", " '113 146': 5.2,\n", " '113 148': 1.1885714285714286,\n", " '113 152': 8.0,\n", " '113 158': 0.9530769230769232,\n", " '113 161': 2.2449999999999997,\n", " '113 162': 2.2125,\n", " '113 163': 2.515,\n", " '113 164': 1.4830769230769232,\n", " '113 17': 4.38,\n", " '113 170': 1.5743749999999999,\n", " '113 181': 5.546666666666667,\n", " '113 186': 1.4091666666666667,\n", " '113 209': 2.893333333333333,\n", " '113 211': 1.0183333333333333,\n", " '113 22': 12.0,\n", " '113 224': 1.4649999999999999,\n", " '113 230': 2.11,\n", " '113 231': 1.7366666666666668,\n", " '113 232': 2.1375,\n", " '113 233': 2.1500000000000004,\n", " '113 234': 0.8635714285714285,\n", " '113 236': 3.9924999999999997,\n", " '113 237': 3.412,\n", " '113 238': 5.324,\n", " '113 239': 5.054,\n", " '113 243': 12.4,\n", " '113 244': 9.4,\n", " '113 246': 2.21,\n", " '113 249': 0.7723076923076924,\n", " '113 255': 3.7,\n", " '113 256': 3.3,\n", " '113 261': 2.0033333333333334,\n", " '113 262': 4.4,\n", " '113 263': 4.0,\n", " '113 264': 0.0,\n", " '113 33': 3.8266666666666667,\n", " '113 36': 6.3,\n", " '113 4': 1.09,\n", " '113 41': 9.2,\n", " '113 42': 8.34,\n", " '113 45': 2.02,\n", " '113 48': 2.2960000000000003,\n", " '113 50': 2.86,\n", " '113 66': 3.2,\n", " '113 68': 1.1981818181818182,\n", " '113 79': 0.7782352941176471,\n", " '113 80': 5.31,\n", " '113 87': 3.135,\n", " '113 88': 2.5999999999999996,\n", " '113 90': 0.8544444444444445,\n", " '113 94': 12.5,\n", " '114 100': 2.3966666666666665,\n", " '114 107': 1.2999999999999998,\n", " '114 112': 5.13,\n", " '114 113': 0.61,\n", " '114 114': 0.5481818181818182,\n", " '114 116': 9.1,\n", " '114 125': 0.752,\n", " '114 13': 1.9000000000000001,\n", " '114 137': 1.8425,\n", " '114 14': 11.11,\n", " '114 140': 4.1,\n", " '114 141': 3.9333333333333336,\n", " '114 142': 3.8766666666666665,\n", " '114 143': 4.745,\n", " '114 144': 0.84875,\n", " '114 145': 6.53,\n", " '114 148': 0.9109090909090909,\n", " '114 151': 7.3,\n", " '114 158': 1.356,\n", " '114 161': 2.80375,\n", " '114 162': 2.7075,\n", " '114 163': 3.5,\n", " '114 164': 1.7079999999999997,\n", " '114 166': 7.55,\n", " '114 169': 11.6,\n", " '114 170': 2.0833333333333335,\n", " '114 181': 3.77,\n", " '114 186': 1.6784615384615384,\n", " '114 190': 4.7,\n", " '114 209': 1.66,\n", " '114 211': 0.45,\n", " '114 217': 2.4,\n", " '114 223': 7.6,\n", " '114 224': 5.23,\n", " '114 225': 4.41,\n", " '114 229': 3.0,\n", " '114 230': 2.8775,\n", " '114 231': 1.2021428571428572,\n", " '114 232': 1.3399999999999999,\n", " '114 233': 2.3375,\n", " '114 234': 1.3515384615384616,\n", " '114 236': 4.5649999999999995,\n", " '114 237': 3.716666666666667,\n", " '114 238': 5.89,\n", " '114 239': 4.99,\n", " '114 24': 6.673333333333333,\n", " '114 243': 11.23,\n", " '114 244': 8.9,\n", " '114 246': 2.0066666666666664,\n", " '114 249': 0.8476923076923076,\n", " '114 255': 3.6879999999999997,\n", " '114 257': 4.96,\n", " '114 260': 7.08,\n", " '114 261': 1.7333333333333334,\n", " '114 262': 5.505,\n", " '114 263': 4.75,\n", " '114 36': 5.98,\n", " '114 4': 1.35,\n", " '114 43': 3.6,\n", " '114 45': 1.0866666666666667,\n", " '114 48': 3.3966666666666665,\n", " '114 49': 3.965,\n", " '114 50': 3.315,\n", " '114 62': 5.7,\n", " '114 65': 2.8,\n", " '114 66': 3.3,\n", " '114 68': 1.7022222222222223,\n", " '114 69': 10.05,\n", " '114 7': 6.4,\n", " '114 79': 1.0336363636363635,\n", " '114 87': 2.035,\n", " '114 90': 1.3199999999999998,\n", " '114 97': 3.7,\n", " '116 116': 0.47800000000000004,\n", " '116 119': 2.9,\n", " '116 132': 19.05,\n", " '116 159': 1.64,\n", " '116 162': 6.1,\n", " '116 166': 1.3275000000000001,\n", " '116 186': 6.42,\n", " '116 230': 6.38,\n", " '116 238': 3.2975000000000003,\n", " '116 239': 4.52,\n", " '116 244': 1.085,\n", " '116 41': 1.7149999999999999,\n", " '116 42': 1.5825,\n", " '116 68': 6.31,\n", " '116 74': 2.08,\n", " '116 75': 4.23,\n", " '116 79': 9.3,\n", " '118 118': 1.43,\n", " '12 100': 4.0,\n", " '12 13': 0.9,\n", " '12 142': 5.56,\n", " '12 144': 2.08,\n", " '12 151': 8.3,\n", " '12 163': 5.5,\n", " '12 164': 5.38,\n", " '12 170': 4.9,\n", " '12 48': 4.67,\n", " '123 123': 0.93,\n", " '125 1': 14.67,\n", " '125 100': 2.1100000000000003,\n", " '125 106': 5.0,\n", " '125 107': 2.1316666666666664,\n", " '125 113': 0.7,\n", " '125 114': 0.825,\n", " '125 129': 8.13,\n", " '125 13': 1.3,\n", " '125 132': 19.88,\n", " '125 137': 2.68,\n", " '125 138': 10.4575,\n", " '125 140': 4.955,\n", " '125 141': 6.5,\n", " '125 142': 4.8,\n", " '125 144': 0.6583333333333333,\n", " '125 148': 1.3885714285714283,\n", " '125 151': 5.43,\n", " '125 158': 0.7375,\n", " '125 161': 2.783333333333333,\n", " '125 162': 3.245,\n", " '125 163': 3.4000000000000004,\n", " '125 164': 2.615,\n", " '125 170': 4.0275,\n", " '125 186': 1.8719999999999999,\n", " '125 188': 5.88,\n", " '125 211': 0.66,\n", " '125 227': 9.4,\n", " '125 230': 2.71,\n", " '125 231': 1.0077777777777779,\n", " '125 234': 1.7242857142857144,\n", " '125 236': 4.8,\n", " '125 237': 3.47,\n", " '125 238': 6.09,\n", " '125 239': 5.05,\n", " '125 244': 9.07,\n", " '125 246': 2.265,\n", " '125 249': 0.6785714285714286,\n", " '125 255': 4.050000000000001,\n", " '125 256': 3.1,\n", " '125 261': 2.01,\n", " '125 263': 5.949999999999999,\n", " '125 42': 8.16,\n", " '125 48': 2.74,\n", " '125 49': 4.21,\n", " '125 68': 1.93,\n", " '125 75': 7.2,\n", " '125 79': 1.5642857142857143,\n", " '125 87': 2.0875,\n", " '125 88': 2.505,\n", " '125 90': 1.435,\n", " '125 97': 4.835,\n", " '127 243': 1.92,\n", " '128 238': 7.3,\n", " '129 129': 0.808,\n", " '129 160': 6.3,\n", " '129 164': 1.96,\n", " '129 173': 2.1,\n", " '129 207': 1.2,\n", " '129 70': 1.69,\n", " '13 100': 3.9799999999999995,\n", " '13 107': 4.665,\n", " '13 113': 2.75,\n", " '13 114': 2.0975,\n", " '13 12': 0.9,\n", " '13 125': 0.93,\n", " '13 13': 0.518,\n", " '13 132': 24.505,\n", " '13 137': 5.045,\n", " '13 138': 15.219999999999999,\n", " '13 14': 7.1,\n", " '13 140': 6.955,\n", " '13 141': 7.4399999999999995,\n", " '13 142': 5.1,\n", " '13 143': 5.15,\n", " '13 144': 2.1500000000000004,\n", " '13 148': 3.3075,\n", " '13 158': 2.35,\n", " '13 161': 5.9030000000000005,\n", " '13 162': 6.3725,\n", " '13 163': 5.172857142857143,\n", " '13 164': 6.02,\n", " '13 166': 7.46,\n", " '13 17': 5.6,\n", " '13 170': 6.166666666666667,\n", " '13 181': 3.9,\n", " '13 186': 3.7733333333333334,\n", " '13 209': 1.8,\n", " '13 211': 1.7325,\n", " '13 224': 4.8,\n", " '13 225': 7.51,\n", " '13 226': 8.3,\n", " '13 229': 6.343999999999999,\n", " '13 230': 4.56,\n", " '13 231': 0.9583333333333334,\n", " '13 232': 3.13,\n", " '13 233': 6.2,\n", " '13 234': 3.8175,\n", " '13 236': 8.33,\n", " '13 237': 7.136666666666667,\n", " '13 238': 6.7,\n", " '13 239': 5.715,\n", " '13 244': 10.575,\n", " '13 246': 2.7916666666666665,\n", " '13 249': 2.145,\n", " '13 25': 3.0,\n", " '13 255': 5.53,\n", " '13 261': 0.78,\n", " '13 262': 8.11,\n", " '13 263': 8.07,\n", " '13 33': 3.99,\n", " '13 40': 2.8,\n", " '13 45': 2.4,\n", " '13 48': 4.154285714285714,\n", " '13 49': 5.43,\n", " '13 50': 4.1,\n", " '13 54': 4.62,\n", " '13 55': 13.26,\n", " '13 65': 3.9600000000000004,\n", " '13 68': 2.7,\n", " '13 74': 10.36,\n", " '13 79': 3.9574999999999996,\n", " '13 85': 10.99,\n", " '13 87': 1.2125,\n", " '13 88': 0.9,\n", " '13 90': 2.7,\n", " '13 91': 7.86,\n", " '130 230': 12.8,\n", " '130 64': 6.03,\n", " '131 9': 2.1,\n", " '132 10': 3.75,\n", " '132 100': 17.6,\n", " '132 102': 7.7,\n", " '132 106': 20.2,\n", " '132 107': 17.561666666666667,\n", " '132 11': 17.945,\n", " '132 112': 15.809999999999999,\n", " '132 113': 18.302,\n", " '132 114': 21.73,\n", " '132 117': 12.2,\n", " '132 121': 10.47,\n", " '132 123': 15.65,\n", " '132 124': 5.67,\n", " '132 125': 18.736666666666668,\n", " '132 13': 20.86,\n", " '132 130': 6.716666666666666,\n", " '132 132': 2.2558620689655173,\n", " '132 134': 6.576666666666667,\n", " '132 137': 16.720000000000002,\n", " '132 138': 11.68625,\n", " '132 14': 20.065,\n", " '132 140': 19.293333333333333,\n", " '132 141': 19.14,\n", " '132 142': 20.406666666666666,\n", " '132 143': 10.905000000000001,\n", " '132 144': 18.537499999999998,\n", " '132 145': 15.837142857142856,\n", " '132 148': 17.994285714285716,\n", " '132 149': 14.32,\n", " '132 15': 14.4,\n", " '132 150': 14.85,\n", " '132 151': 19.834,\n", " '132 152': 19.1,\n", " '132 158': 22.7,\n", " '132 161': 18.601666666666667,\n", " '132 162': 17.082857142857144,\n", " '132 163': 19.229,\n", " '132 164': 18.7575,\n", " '132 166': 18.6,\n", " '132 17': 10.4,\n", " '132 170': 17.203,\n", " '132 174': 21.17,\n", " '132 177': 9.2,\n", " '132 179': 15.27,\n", " '132 181': 17.358571428571427,\n", " '132 186': 18.375,\n", " '132 188': 12.149999999999999,\n", " '132 189': 12.2,\n", " '132 19': 10.5,\n", " '132 195': 26.54,\n", " '132 196': 9.35,\n", " '132 197': 6.59,\n", " '132 198': 9.9,\n", " '132 201': 12.94,\n", " '132 205': 6.0,\n", " '132 209': 21.2,\n", " '132 211': 18.91,\n", " '132 212': 16.85,\n", " '132 213': 15.2,\n", " '132 215': 4.9,\n", " '132 216': 4.487,\n", " '132 218': 4.5,\n", " '132 22': 17.9,\n", " '132 220': 30.5,\n", " '132 222': 8.21,\n", " '132 223': 13.25,\n", " '132 224': 17.59,\n", " '132 225': 11.8,\n", " '132 226': 14.959999999999999,\n", " '132 228': 23.875,\n", " '132 229': 18.49,\n", " '132 23': 30.83,\n", " '132 230': 18.5712,\n", " '132 231': 20.46,\n", " '132 232': 18.3,\n", " '132 233': 17.86,\n", " '132 234': 17.654,\n", " '132 236': 19.491666666666667,\n", " '132 237': 19.540000000000003,\n", " '132 238': 20.8375,\n", " '132 239': 20.90125,\n", " '132 24': 19.14,\n", " '132 241': 20.5,\n", " '132 243': 22.1,\n", " '132 244': 19.9,\n", " '132 246': 18.515,\n", " '132 248': 17.22,\n", " '132 249': 18.7325,\n", " '132 25': 14.808000000000002,\n", " '132 252': 11.3,\n", " '132 255': 16.466666666666665,\n", " '132 256': 17.224,\n", " '132 257': 19.81,\n", " '132 259': 20.96,\n", " '132 26': 13.92,\n", " '132 261': 22.115000000000002,\n", " '132 262': 19.165,\n", " '132 263': 19.21,\n", " '132 264': 0.0,\n", " '132 265': 14.885833333333332,\n", " '132 28': 6.3133333333333335,\n", " '132 33': 18.683333333333334,\n", " '132 36': 15.886666666666665,\n", " '132 37': 14.899999999999999,\n", " '132 38': 7.300000000000001,\n", " '132 39': 9.912857142857144,\n", " '132 4': 18.59,\n", " '132 40': 14.1,\n", " '132 42': 17.95,\n", " '132 43': 18.744999999999997,\n", " '132 48': 18.761904761904763,\n", " '132 49': 11.92,\n", " '132 50': 18.735,\n", " '132 51': 19.064999999999998,\n", " '132 52': 26.86,\n", " '132 54': 27.2,\n", " '132 55': 17.3,\n", " '132 61': 10.69,\n", " '132 62': 13.229999999999999,\n", " '132 64': 13.6,\n", " '132 65': 15.686000000000002,\n", " '132 66': 19.3,\n", " '132 68': 18.7975,\n", " '132 7': 14.783333333333333,\n", " '132 70': 11.3,\n", " '132 71': 11.34,\n", " '132 72': 10.19,\n", " '132 74': 17.25,\n", " '132 76': 9.256666666666666,\n", " '132 77': 9.0,\n", " '132 79': 19.43166666666667,\n", " '132 80': 15.606666666666667,\n", " '132 82': 10.335,\n", " '132 83': 11.15,\n", " '132 85': 13.45,\n", " '132 86': 7.8,\n", " '132 87': 19.96,\n", " '132 88': 20.6,\n", " '132 89': 14.71,\n", " '132 9': 16.51,\n", " '132 90': 18.666666666666668,\n", " '132 91': 13.835,\n", " '132 92': 10.515,\n", " '132 93': 10.265,\n", " '132 95': 8.1,\n", " '132 97': 16.35,\n", " '133 133': 4.43,\n", " '134 197': 2.2,\n", " '135 75': 12.85,\n", " '137 100': 1.458,\n", " '137 107': 0.6836363636363636,\n", " '137 112': 5.1,\n", " '137 113': 1.338,\n", " '137 114': 1.7600000000000002,\n", " '137 125': 2.33,\n", " '137 13': 5.48,\n", " '137 132': 22.26,\n", " '137 135': 10.36,\n", " '137 137': 0.4633333333333333,\n", " '137 138': 8.4,\n", " '137 14': 12.34,\n", " '137 140': 2.186666666666667,\n", " '137 141': 1.83,\n", " '137 142': 3.08,\n", " '137 145': 2.7,\n", " '137 148': 1.4,\n", " '137 158': 2.66,\n", " '137 161': 1.46875,\n", " '137 162': 1.1652173913043478,\n", " '137 163': 1.99,\n", " '137 164': 0.6784615384615384,\n", " '137 170': 0.7204347826086956,\n", " '137 181': 7.65,\n", " '137 186': 0.9963636363636365,\n", " '137 209': 4.09,\n", " '137 220': 12.3,\n", " '137 223': 7.03,\n", " '137 224': 0.9,\n", " '137 229': 1.264,\n", " '137 230': 1.5200000000000002,\n", " '137 231': 3.3175,\n", " '137 232': 2.5949999999999998,\n", " '137 233': 0.886875,\n", " '137 234': 1.0316666666666667,\n", " '137 236': 3.0,\n", " '137 237': 2.2125,\n", " '137 238': 5.4,\n", " '137 239': 4.43,\n", " '137 243': 9.69,\n", " '137 246': 2.11,\n", " '137 249': 2.1574999999999998,\n", " '137 255': 4.35,\n", " '137 261': 5.49,\n", " '137 262': 3.005,\n", " '137 263': 3.5325,\n", " '137 4': 1.6219999999999999,\n", " '137 41': 5.125,\n", " '137 42': 6.91,\n", " '137 43': 2.9,\n", " '137 45': 2.36,\n", " '137 48': 2.1533333333333333,\n", " '137 50': 2.7,\n", " '137 61': 8.133333333333333,\n", " '137 68': 1.672,\n", " '137 7': 4.2,\n", " '137 74': 4.68,\n", " '137 79': 1.30875,\n", " '137 82': 5.8,\n", " '137 87': 4.34,\n", " '137 88': 4.449999999999999,\n", " '137 90': 1.3,\n", " '138 1': 32.72,\n", " '138 100': 9.765,\n", " '138 106': 11.0,\n", " '138 107': 9.463333333333333,\n", " '138 112': 7.25,\n", " '138 113': 11.0875,\n", " '138 114': 11.45,\n", " '138 116': 8.015,\n", " '138 121': 6.7,\n", " '138 125': 14.567499999999999,\n", " '138 127': 10.16,\n", " '138 129': 4.00875,\n", " '138 13': 14.410000000000002,\n", " '138 130': 7.21,\n", " '138 132': 12.577142857142858,\n", " '138 134': 6.859999999999999,\n", " '138 137': 8.75,\n", " '138 138': 0.9528571428571428,\n", " '138 14': 16.18,\n", " '138 140': 8.88125,\n", " '138 141': 9.37,\n", " '138 142': 10.852307692307694,\n", " '138 143': 10.466666666666667,\n", " '138 144': 12.341666666666667,\n", " '138 145': 7.723333333333333,\n", " '138 146': 4.134,\n", " '138 148': 11.54,\n", " '138 15': 8.100000000000001,\n", " '138 151': 9.15,\n", " '138 152': 8.86,\n", " '138 158': 13.899999999999999,\n", " '138 160': 6.68,\n", " '138 161': 10.131388888888889,\n", " '138 162': 9.673,\n", " '138 163': 10.425714285714285,\n", " '138 164': 9.65157894736842,\n", " '138 166': 8.26,\n", " '138 17': 9.2425,\n", " '138 170': 8.977058823529413,\n", " '138 171': 7.33,\n", " '138 174': 14.1,\n", " '138 175': 9.3,\n", " '138 177': 10.95,\n", " '138 178': 18.225,\n", " '138 179': 3.9266666666666663,\n", " '138 180': 11.2,\n", " '138 181': 11.776666666666666,\n", " '138 182': 10.29,\n", " '138 186': 11.034,\n", " '138 188': 13.36,\n", " '138 189': 13.433333333333332,\n", " '138 192': 4.66,\n", " '138 196': 5.6066666666666665,\n", " '138 197': 7.495,\n", " '138 198': 9.96,\n", " '138 200': 12.3,\n", " '138 209': 12.950000000000001,\n", " '138 210': 20.5,\n", " '138 211': 12.796666666666667,\n", " '138 220': 13.39,\n", " '138 223': 3.0500000000000003,\n", " '138 224': 9.870000000000001,\n", " '138 225': 9.817499999999999,\n", " '138 226': 4.58,\n", " '138 229': 10.012,\n", " '138 230': 10.601590909090909,\n", " '138 231': 13.138333333333334,\n", " '138 232': 11.485,\n", " '138 233': 8.775555555555556,\n", " '138 234': 10.5275,\n", " '138 236': 8.83,\n", " '138 237': 9.46625,\n", " '138 238': 9.202222222222222,\n", " '138 239': 10.172727272727274,\n", " '138 243': 10.42,\n", " '138 244': 10.085,\n", " '138 246': 10.52,\n", " '138 249': 11.162,\n", " '138 25': 10.888333333333334,\n", " '138 252': 5.23,\n", " '138 255': 7.432222222222222,\n", " '138 256': 8.68,\n", " '138 257': 16.4,\n", " '138 260': 4.27,\n", " '138 261': 16.1,\n", " '138 262': 8.055,\n", " '138 263': 8.690000000000001,\n", " '138 265': 20.552,\n", " '138 29': 21.65,\n", " '138 33': 11.03,\n", " '138 36': 7.7299999999999995,\n", " '138 37': 8.565,\n", " '138 4': 10.49,\n", " '138 41': 8.065,\n", " '138 42': 7.01,\n", " '138 43': 10.445,\n", " '138 48': 10.377692307692307,\n", " '138 49': 9.823333333333332,\n", " '138 50': 10.78,\n", " '138 51': 13.8,\n", " '138 52': 12.3,\n", " '138 53': 5.3,\n", " '138 56': 4.1,\n", " '138 61': 13.31,\n", " '138 62': 10.11,\n", " '138 64': 12.36,\n", " '138 65': 9.932857142857143,\n", " '138 66': 10.6,\n", " '138 68': 11.475000000000001,\n", " '138 69': 7.5,\n", " '138 7': 3.61125,\n", " '138 70': 1.4266666666666667,\n", " '138 74': 6.7875,\n", " '138 75': 8.1475,\n", " '138 79': 10.967500000000001,\n", " '138 80': 6.99,\n", " '138 81': 14.47,\n", " '138 82': 3.52,\n", " '138 83': 3.335,\n", " '138 87': 13.8125,\n", " '138 88': 15.393333333333333,\n", " '138 89': 15.11,\n", " '138 90': 10.943999999999999,\n", " '138 92': 3.65,\n", " '138 93': 4.29,\n", " '138 95': 5.085,\n", " '138 97': 11.11,\n", " '138 98': 8.7,\n", " '14 14': 0.38,\n", " '140 107': 3.115,\n", " '140 113': 4.1975,\n", " '140 125': 8.28,\n", " '140 13': 7.64,\n", " '140 132': 18.7,\n", " '140 135': 14.48,\n", " '140 137': 2.58,\n", " '140 138': 9.305,\n", " '140 140': 0.5986363636363636,\n", " '140 141': 0.7527272727272727,\n", " '140 142': 1.7822222222222222,\n", " '140 143': 2.32,\n", " '140 151': 2.83,\n", " '140 161': 1.8414285714285714,\n", " '140 162': 1.4966666666666666,\n", " '140 163': 1.5966666666666667,\n", " '140 164': 2.4683333333333333,\n", " '140 166': 4.2,\n", " '140 170': 2.122,\n", " '140 179': 4.22,\n", " '140 186': 3.31625,\n", " '140 193': 4.15,\n", " '140 209': 6.17,\n", " '140 211': 5.53,\n", " '140 223': 6.4,\n", " '140 224': 3.1,\n", " '140 226': 3.62,\n", " '140 229': 1.1131578947368421,\n", " '140 230': 2.41,\n", " '140 231': 7.45,\n", " '140 232': 5.75,\n", " '140 233': 1.7408333333333335,\n", " '140 234': 3.5033333333333334,\n", " '140 236': 1.2205714285714286,\n", " '140 237': 0.9748837209302326,\n", " '140 238': 2.196,\n", " '140 239': 2.347142857142857,\n", " '140 24': 4.36,\n", " '140 243': 6.84,\n", " '140 244': 7.56,\n", " '140 246': 4.95,\n", " '140 249': 5.073333333333333,\n", " '140 260': 4.82,\n", " '140 262': 0.8657692307692308,\n", " '140 263': 0.9194117647058824,\n", " '140 4': 3.825,\n", " '140 43': 1.7,\n", " '140 45': 5.0,\n", " '140 48': 2.669166666666667,\n", " '140 50': 3.0,\n", " '140 52': 7.66,\n", " '140 65': 7.12,\n", " '140 66': 7.5,\n", " '140 68': 4.1025,\n", " '140 7': 4.96,\n", " '140 74': 2.9725,\n", " '140 75': 1.807142857142857,\n", " '140 79': 4.2,\n", " '140 83': 5.7,\n", " '140 85': 11.32,\n", " '140 87': 5.8933333333333335,\n", " '140 88': 6.109999999999999,\n", " '140 90': 3.685,\n", " '140 95': 7.6,\n", " '140 97': 9.0,\n", " '141 100': 2.5149999999999997,\n", " '141 107': 2.60125,\n", " '141 112': 4.4,\n", " '141 113': 3.75,\n", " '141 114': 3.9,\n", " '141 116': 6.41,\n", " '141 13': 7.11,\n", " '141 130': 13.8,\n", " '141 132': 19.41,\n", " '141 133': 11.54,\n", " '141 137': 2.312307692307692,\n", " '141 138': 9.285,\n", " '141 140': 0.9331578947368421,\n", " '141 141': 0.8211764705882354,\n", " '141 142': 1.7133333333333332,\n", " '141 143': 1.99,\n", " '141 145': 2.9,\n", " '141 148': 5.65,\n", " '141 151': 3.125,\n", " '141 158': 4.8,\n", " '141 161': 1.5977777777777777,\n", " '141 162': 1.142121212121212,\n", " '141 163': 1.2077777777777778,\n", " '141 164': 2.2333333333333334,\n", " '141 166': 4.166666666666667,\n", " '141 170': 1.7866666666666668,\n", " '141 173': 6.45,\n", " '141 178': 14.0,\n", " '141 186': 2.7445454545454546,\n", " '141 193': 2.72,\n", " '141 196': 7.63,\n", " '141 209': 6.66,\n", " '141 211': 5.2,\n", " '141 220': 10.23,\n", " '141 224': 3.0,\n", " '141 226': 2.48,\n", " '141 229': 0.9359090909090909,\n", " '141 230': 1.8555555555555554,\n", " '141 231': 7.246666666666666,\n", " '141 233': 1.3258333333333334,\n", " '141 234': 2.982,\n", " '141 236': 1.1402083333333333,\n", " '141 237': 0.6136842105263158,\n", " '141 238': 2.375,\n", " '141 239': 2.0327272727272727,\n", " '141 24': 3.9,\n", " '141 243': 7.63,\n", " '141 244': 7.68,\n", " '141 246': 4.55,\n", " '141 249': 5.66,\n", " '141 255': 5.13,\n", " '141 261': 6.885,\n", " '141 262': 0.8171428571428571,\n", " '141 263': 0.9044444444444445,\n", " '141 4': 4.6675,\n", " '141 42': 4.046666666666667,\n", " '141 43': 1.231111111111111,\n", " '141 48': 2.526666666666667,\n", " '141 50': 2.6775,\n", " '141 65': 8.0,\n", " '141 68': 3.1975,\n", " '141 7': 3.5333333333333337,\n", " '141 74': 3.1374999999999997,\n", " '141 75': 1.90125,\n", " '141 79': 3.9475,\n", " '141 80': 5.5649999999999995,\n", " '141 88': 7.26,\n", " '141 90': 4.05,\n", " '142 100': 1.6228571428571428,\n", " '142 107': 3.2199999999999998,\n", " '142 113': 3.2,\n", " '142 114': 3.74,\n", " '142 116': 4.556666666666667,\n", " '142 125': 3.99,\n", " '142 127': 8.965,\n", " '142 129': 5.68,\n", " '142 13': 5.058333333333334,\n", " '142 132': 20.77,\n", " '142 137': 2.985,\n", " '142 138': 9.133333333333333,\n", " '142 140': 2.2916666666666665,\n", " '142 141': 1.7023076923076923,\n", " '142 142': 0.628974358974359,\n", " '142 143': 0.8415384615384615,\n", " '142 144': 4.48,\n", " '142 145': 3.6,\n", " '142 148': 7.87,\n", " '142 151': 2.0314285714285716,\n", " '142 158': 2.7,\n", " '142 161': 1.426875,\n", " '142 162': 1.6821052631578948,\n", " '142 163': 0.8309375,\n", " '142 164': 2.382,\n", " '142 166': 2.6875,\n", " '142 17': 8.274999999999999,\n", " '142 170': 2.3214285714285716,\n", " '142 174': 12.6,\n", " '142 181': 8.14,\n", " '142 186': 1.8606666666666667,\n", " '142 209': 7.3,\n", " '142 211': 5.0,\n", " '142 220': 9.0,\n", " '142 223': 5.795,\n", " '142 224': 3.885,\n", " '142 225': 8.8,\n", " '142 229': 1.6320000000000001,\n", " '142 230': 1.051212121212121,\n", " '142 231': 4.8425,\n", " '142 233': 2.2944444444444443,\n", " '142 234': 2.9166666666666665,\n", " '142 236': 2.0282758620689654,\n", " '142 237': 1.3573333333333333,\n", " '142 238': 1.44875,\n", " '142 239': 0.9964999999999999,\n", " '142 24': 2.1580000000000004,\n", " '142 243': 7.75,\n", " '142 244': 6.058333333333334,\n", " '142 246': 2.075714285714286,\n", " '142 249': 2.982,\n", " '142 261': 6.45,\n", " '142 262': 2.686666666666667,\n", " '142 263': 2.29,\n", " '142 264': 0.4,\n", " '142 41': 2.9244444444444446,\n", " '142 42': 3.94,\n", " '142 43': 1.1046153846153848,\n", " '142 48': 0.9956756756756756,\n", " '142 50': 1.0758333333333334,\n", " '142 68': 1.8776470588235294,\n", " '142 74': 3.8925,\n", " ...}" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# 1. Convert `grouped` to a dictionary\n", "### YOUR CODE HERE ###\n", "grouped_dict = grouped.to_dict()\n", "# 2. Reassign to only contain the inner dictionary\n", "### YOUR CODE HERE ###\n", "grouped_dict" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "1. Create a `mean_distance` column that is a copy of the `pickup_dropoff` helper column.\n", "\n", "2. Use the [`map()`](https://pandas.pydata.org/docs/reference/api/pandas.Series.map.html#pandas-series-map) method on the `mean_distance` series. Pass `grouped_dict` as its argument. Reassign the result back to the `mean_distance` series.\n", "

\n", "When you pass a dictionary to the `Series.map()` method, it will replace the data in the series where that data matches the dictionary's keys. The values that get imputed are the values of the dictionary.\n", "\n", "```\n", "Example:\n", "df['mean_distance']\n", "```\n", "\n", "|mean_distance |\n", "| :-: |\n", "| 'A B' |\n", "| 'C D' |\n", "| 'A B' |\n", "| 'D C' |\n", "| 'E F' |\n", "\n", "```\n", "grouped_dict = {'A B': 1.25, 'C D': 2, 'D C': 3}\n", "df['mean_distance`] = df['mean_distance'].map(grouped_dict)\n", "df['mean_distance']\n", "```\n", "\n", "|mean_distance |\n", "| :-: |\n", "| 1.25 |\n", "| 2 |\n", "| 1.25 |\n", "| 3 |\n", "| NaN |\n", "\n", "When used this way, the `map()` `Series` method is very similar to `replace()`, however, note that `map()` will impute `NaN` for any values in the series that do not have a corresponding key in the mapping dictionary, so be careful." ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0VendorIDtpep_pickup_datetimetpep_dropoff_datetimepassenger_counttrip_distanceRatecodeIDstore_and_fwd_flagPULocationIDDOLocationID...fare_amountextramta_taxtip_amounttolls_amountimprovement_surchargetotal_amountdurationpickup_dropoffmean_distance
02487011422017-03-25 08:55:432017-03-25 09:09:4763.341N100231...13.00.00.52.760.00.316.5614.066667100 2313.521667
13563424912017-04-11 14:53:282017-04-11 15:19:5811.801N18643...16.00.00.54.000.00.320.8026.500000186 433.108889
210620369012017-12-15 07:26:562017-12-15 07:34:0811.001N262236...6.50.00.51.450.00.38.757.200000262 2360.881429
33894213622017-05-07 13:17:592017-05-07 13:48:1413.701N18897...20.50.00.56.390.00.327.6930.250000188 973.700000
43084167022017-04-15 23:32:202017-04-15 23:49:0314.371N4112...16.50.50.50.000.00.317.8016.7166674 1124.435000
\n", "

5 rows × 21 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 VendorID tpep_pickup_datetime tpep_dropoff_datetime \\\n", "0 24870114 2 2017-03-25 08:55:43 2017-03-25 09:09:47 \n", "1 35634249 1 2017-04-11 14:53:28 2017-04-11 15:19:58 \n", "2 106203690 1 2017-12-15 07:26:56 2017-12-15 07:34:08 \n", "3 38942136 2 2017-05-07 13:17:59 2017-05-07 13:48:14 \n", "4 30841670 2 2017-04-15 23:32:20 2017-04-15 23:49:03 \n", "\n", " passenger_count trip_distance RatecodeID store_and_fwd_flag \\\n", "0 6 3.34 1 N \n", "1 1 1.80 1 N \n", "2 1 1.00 1 N \n", "3 1 3.70 1 N \n", "4 1 4.37 1 N \n", "\n", " PULocationID DOLocationID ... fare_amount extra mta_tax tip_amount \\\n", "0 100 231 ... 13.0 0.0 0.5 2.76 \n", "1 186 43 ... 16.0 0.0 0.5 4.00 \n", "2 262 236 ... 6.5 0.0 0.5 1.45 \n", "3 188 97 ... 20.5 0.0 0.5 6.39 \n", "4 4 112 ... 16.5 0.5 0.5 0.00 \n", "\n", " tolls_amount improvement_surcharge total_amount duration \\\n", "0 0.0 0.3 16.56 14.066667 \n", "1 0.0 0.3 20.80 26.500000 \n", "2 0.0 0.3 8.75 7.200000 \n", "3 0.0 0.3 27.69 30.250000 \n", "4 0.0 0.3 17.80 16.716667 \n", "\n", " pickup_dropoff mean_distance \n", "0 100 231 3.521667 \n", "1 186 43 3.108889 \n", "2 262 236 0.881429 \n", "3 188 97 3.700000 \n", "4 4 112 4.435000 \n", "\n", "[5 rows x 21 columns]" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# 1. Create a mean_distance column that is a copy of the pickup_dropoff helper column\n", "### YOUR CODE HERE ###\n", "df0['mean_distance'] = df0['pickup_dropoff']\n", "# 2. Map `grouped_dict` to the `mean_distance` column\n", "### YOUR CODE HERE ###\n", "df0['mean_distance'] = df0['mean_distance'].map(grouped_dict)\n", "# Confirm that it worked\n", "### YOUR CODE HERE ###\n", "df0.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Create `mean_duration` column\n", "\n", "Repeat the process used to create the `mean_distance` column to create a `mean_duration` column." ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0VendorIDtpep_pickup_datetimetpep_dropoff_datetimepassenger_counttrip_distanceRatecodeIDstore_and_fwd_flagPULocationIDDOLocationID...extramta_taxtip_amounttolls_amountimprovement_surchargetotal_amountdurationpickup_dropoffmean_distancemean_duration
02487011422017-03-25 08:55:432017-03-25 09:09:4763.341N100231...0.00.52.760.00.316.5614.066667100 2313.52166722.847222
13563424912017-04-11 14:53:282017-04-11 15:19:5811.801N18643...0.00.54.000.00.320.8026.500000186 433.10888924.470370
210620369012017-12-15 07:26:562017-12-15 07:34:0811.001N262236...0.00.51.450.00.38.757.200000262 2360.8814297.250000
33894213622017-05-07 13:17:592017-05-07 13:48:1413.701N18897...0.00.56.390.00.327.6930.250000188 973.70000030.250000
43084167022017-04-15 23:32:202017-04-15 23:49:0314.371N4112...0.50.50.000.00.317.8016.7166674 1124.43500014.616667
\n", "

5 rows × 22 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 VendorID tpep_pickup_datetime tpep_dropoff_datetime \\\n", "0 24870114 2 2017-03-25 08:55:43 2017-03-25 09:09:47 \n", "1 35634249 1 2017-04-11 14:53:28 2017-04-11 15:19:58 \n", "2 106203690 1 2017-12-15 07:26:56 2017-12-15 07:34:08 \n", "3 38942136 2 2017-05-07 13:17:59 2017-05-07 13:48:14 \n", "4 30841670 2 2017-04-15 23:32:20 2017-04-15 23:49:03 \n", "\n", " passenger_count trip_distance RatecodeID store_and_fwd_flag \\\n", "0 6 3.34 1 N \n", "1 1 1.80 1 N \n", "2 1 1.00 1 N \n", "3 1 3.70 1 N \n", "4 1 4.37 1 N \n", "\n", " PULocationID DOLocationID ... extra mta_tax tip_amount tolls_amount \\\n", "0 100 231 ... 0.0 0.5 2.76 0.0 \n", "1 186 43 ... 0.0 0.5 4.00 0.0 \n", "2 262 236 ... 0.0 0.5 1.45 0.0 \n", "3 188 97 ... 0.0 0.5 6.39 0.0 \n", "4 4 112 ... 0.5 0.5 0.00 0.0 \n", "\n", " improvement_surcharge total_amount duration pickup_dropoff \\\n", "0 0.3 16.56 14.066667 100 231 \n", "1 0.3 20.80 26.500000 186 43 \n", "2 0.3 8.75 7.200000 262 236 \n", "3 0.3 27.69 30.250000 188 97 \n", "4 0.3 17.80 16.716667 4 112 \n", "\n", " mean_distance mean_duration \n", "0 3.521667 22.847222 \n", "1 3.108889 24.470370 \n", "2 0.881429 7.250000 \n", "3 3.700000 30.250000 \n", "4 4.435000 14.616667 \n", "\n", "[5 rows x 22 columns]" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "### YOUR CODE HERE ###\n", "grouped_duration = df0.groupby('pickup_dropoff')['duration'].mean()\n", "grouped_duration = grouped_duration.to_dict()\n", "\n", "# Create a dictionary where keys are unique pickup_dropoffs and values are\n", "# mean trip duration for all trips with those pickup_dropoff combos\n", "### YOUR CODE HERE ###\n", "df0['mean_duration'] = df0['pickup_dropoff']\n", "df0['mean_duration'] = df0['mean_duration'].map(grouped_duration)\n", "# Confirm that it worked\n", "### YOUR CODE HERE ###\n", "df0.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Create `day` and `month` columns\n", "\n", "Create two new columns, `day` (name of day) and `month` (name of month) by extracting the relevant information from the `tpep_pickup_datetime` column." ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0VendorIDtpep_pickup_datetimetpep_dropoff_datetimepassenger_counttrip_distanceRatecodeIDstore_and_fwd_flagPULocationIDDOLocationID...tip_amounttolls_amountimprovement_surchargetotal_amountdurationpickup_dropoffmean_distancemean_durationdaymonth
02487011422017-03-25 08:55:432017-03-25 09:09:4763.341N100231...2.760.00.316.5614.066667100 2313.52166722.847222SaturdayMarch
13563424912017-04-11 14:53:282017-04-11 15:19:5811.801N18643...4.000.00.320.8026.500000186 433.10888924.470370TuesdayApril
210620369012017-12-15 07:26:562017-12-15 07:34:0811.001N262236...1.450.00.38.757.200000262 2360.8814297.250000FridayDecember
33894213622017-05-07 13:17:592017-05-07 13:48:1413.701N18897...6.390.00.327.6930.250000188 973.70000030.250000SundayMay
43084167022017-04-15 23:32:202017-04-15 23:49:0314.371N4112...0.000.00.317.8016.7166674 1124.43500014.616667SaturdayApril
\n", "

5 rows × 24 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 VendorID tpep_pickup_datetime tpep_dropoff_datetime \\\n", "0 24870114 2 2017-03-25 08:55:43 2017-03-25 09:09:47 \n", "1 35634249 1 2017-04-11 14:53:28 2017-04-11 15:19:58 \n", "2 106203690 1 2017-12-15 07:26:56 2017-12-15 07:34:08 \n", "3 38942136 2 2017-05-07 13:17:59 2017-05-07 13:48:14 \n", "4 30841670 2 2017-04-15 23:32:20 2017-04-15 23:49:03 \n", "\n", " passenger_count trip_distance RatecodeID store_and_fwd_flag \\\n", "0 6 3.34 1 N \n", "1 1 1.80 1 N \n", "2 1 1.00 1 N \n", "3 1 3.70 1 N \n", "4 1 4.37 1 N \n", "\n", " PULocationID DOLocationID ... tip_amount tolls_amount \\\n", "0 100 231 ... 2.76 0.0 \n", "1 186 43 ... 4.00 0.0 \n", "2 262 236 ... 1.45 0.0 \n", "3 188 97 ... 6.39 0.0 \n", "4 4 112 ... 0.00 0.0 \n", "\n", " improvement_surcharge total_amount duration pickup_dropoff \\\n", "0 0.3 16.56 14.066667 100 231 \n", "1 0.3 20.80 26.500000 186 43 \n", "2 0.3 8.75 7.200000 262 236 \n", "3 0.3 27.69 30.250000 188 97 \n", "4 0.3 17.80 16.716667 4 112 \n", "\n", " mean_distance mean_duration day month \n", "0 3.521667 22.847222 Saturday March \n", "1 3.108889 24.470370 Tuesday April \n", "2 0.881429 7.250000 Friday December \n", "3 3.700000 30.250000 Sunday May \n", "4 4.435000 14.616667 Saturday April \n", "\n", "[5 rows x 24 columns]" ] }, "execution_count": 30, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Create 'day' col\n", "### YOUR CODE HERE ###\n", "df0['day'] = df0['tpep_pickup_datetime'].dt.day_name()\n", "# Create 'month' col\n", "### YOUR CODE HERE ###\n", "df0['month'] = df0['tpep_pickup_datetime'].dt.month_name()\n", "\n", "df0.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Create `rush_hour` column\n", "\n", "Define rush hour as:\n", "* Any weekday (not Saturday or Sunday) AND\n", "* Either from 06:00–10:00 or from 16:00–20:00\n", "\n", "Create a binary `rush_hour` column that contains a 1 if the ride was during rush hour and a 0 if it was not." ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0VendorIDtpep_pickup_datetimetpep_dropoff_datetimepassenger_counttrip_distanceRatecodeIDstore_and_fwd_flagPULocationIDDOLocationID...tolls_amountimprovement_surchargetotal_amountdurationpickup_dropoffmean_distancemean_durationdaymonthrush_hour
02487011422017-03-25 08:55:432017-03-25 09:09:4763.341N100231...0.000.316.5614.066667100 2313.52166722.847222SaturdayMarch0
13563424912017-04-11 14:53:282017-04-11 15:19:5811.801N18643...0.000.320.8026.500000186 433.10888924.470370TuesdayApril0
210620369012017-12-15 07:26:562017-12-15 07:34:0811.001N262236...0.000.38.757.200000262 2360.8814297.250000FridayDecember1
33894213622017-05-07 13:17:592017-05-07 13:48:1413.701N18897...0.000.327.6930.250000188 973.70000030.250000SundayMay0
43084167022017-04-15 23:32:202017-04-15 23:49:0314.371N4112...0.000.317.8016.7166674 1124.43500014.616667SaturdayApril0
52334580922017-03-25 20:34:112017-03-25 20:42:1162.301N161236...0.000.312.368.000000161 2362.05225811.855376SaturdayMarch0
63766048722017-05-03 19:04:092017-05-03 20:03:47112.831N79241...0.000.359.1659.63333379 24112.83000059.633333WednesdayMay1
76905941122017-08-15 17:41:062017-08-15 18:03:0512.981N237114...0.000.319.5821.983333237 1144.02250026.437500TuesdayAugust1
8843315922017-02-04 16:17:072017-02-04 16:29:1411.201N234249...0.000.39.8012.116667234 2491.0192597.873457SaturdayFebruary0
99529481712017-11-10 15:20:292017-11-10 15:40:5511.601N239237...0.000.316.5520.433333239 2371.58000010.541111FridayNovember0
101801790922017-03-04 11:58:002017-03-04 12:13:1211.771N162142...0.000.314.7615.200000162 1421.64100014.178333SaturdayMarch0
111860005922017-03-05 19:15:302017-03-05 19:52:18218.902N236132...5.540.372.9236.800000236 13219.21166740.500000SundayMarch0
124678224812017-06-09 19:00:262017-06-09 19:20:1113.001N13148...0.000.320.1519.75000013 1483.30750015.058333FridayJune1
139411324722017-11-06 23:35:052017-11-06 23:42:5712.391N20925...0.000.312.967.866667209 252.3900007.866667MondayNovember0
141416827912017-02-22 15:18:312017-02-22 15:42:5013.301N238161...0.000.322.8524.316667238 1612.93000019.555000WednesdayFebruary0
154744440122017-06-02 06:41:392017-06-02 06:57:4715.931N239231...0.000.322.8016.133333239 2315.95000017.166667FridayJune1
166908867612017-08-15 19:48:082017-08-15 20:00:3713.601N16341...0.000.317.1512.483333163 413.26666714.733333TuesdayAugust1
175869151322017-07-10 13:36:312017-07-10 13:48:4321.711N142100...0.000.310.3012.200000142 1001.62285712.304762MondayJuly0
183538882822017-04-10 18:12:582017-04-10 18:17:3920.631N263262...0.000.36.804.683333263 2620.6621434.577381MondayApril1
191838321422017-03-05 04:01:072017-03-05 04:14:1122.771N7968...0.000.316.0013.06666779 682.13833313.638889SundayMarch0
\n", "

20 rows × 25 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 VendorID tpep_pickup_datetime tpep_dropoff_datetime \\\n", "0 24870114 2 2017-03-25 08:55:43 2017-03-25 09:09:47 \n", "1 35634249 1 2017-04-11 14:53:28 2017-04-11 15:19:58 \n", "2 106203690 1 2017-12-15 07:26:56 2017-12-15 07:34:08 \n", "3 38942136 2 2017-05-07 13:17:59 2017-05-07 13:48:14 \n", "4 30841670 2 2017-04-15 23:32:20 2017-04-15 23:49:03 \n", "5 23345809 2 2017-03-25 20:34:11 2017-03-25 20:42:11 \n", "6 37660487 2 2017-05-03 19:04:09 2017-05-03 20:03:47 \n", "7 69059411 2 2017-08-15 17:41:06 2017-08-15 18:03:05 \n", "8 8433159 2 2017-02-04 16:17:07 2017-02-04 16:29:14 \n", "9 95294817 1 2017-11-10 15:20:29 2017-11-10 15:40:55 \n", "10 18017909 2 2017-03-04 11:58:00 2017-03-04 12:13:12 \n", "11 18600059 2 2017-03-05 19:15:30 2017-03-05 19:52:18 \n", "12 46782248 1 2017-06-09 19:00:26 2017-06-09 19:20:11 \n", "13 94113247 2 2017-11-06 23:35:05 2017-11-06 23:42:57 \n", "14 14168279 1 2017-02-22 15:18:31 2017-02-22 15:42:50 \n", "15 47444401 2 2017-06-02 06:41:39 2017-06-02 06:57:47 \n", "16 69088676 1 2017-08-15 19:48:08 2017-08-15 20:00:37 \n", "17 58691513 2 2017-07-10 13:36:31 2017-07-10 13:48:43 \n", "18 35388828 2 2017-04-10 18:12:58 2017-04-10 18:17:39 \n", "19 18383214 2 2017-03-05 04:01:07 2017-03-05 04:14:11 \n", "\n", " passenger_count trip_distance RatecodeID store_and_fwd_flag \\\n", "0 6 3.34 1 N \n", "1 1 1.80 1 N \n", "2 1 1.00 1 N \n", "3 1 3.70 1 N \n", "4 1 4.37 1 N \n", "5 6 2.30 1 N \n", "6 1 12.83 1 N \n", "7 1 2.98 1 N \n", "8 1 1.20 1 N \n", "9 1 1.60 1 N \n", "10 1 1.77 1 N \n", "11 2 18.90 2 N \n", "12 1 3.00 1 N \n", "13 1 2.39 1 N \n", "14 1 3.30 1 N \n", "15 1 5.93 1 N \n", "16 1 3.60 1 N \n", "17 2 1.71 1 N \n", "18 2 0.63 1 N \n", "19 2 2.77 1 N \n", "\n", " PULocationID DOLocationID ... tolls_amount improvement_surcharge \\\n", "0 100 231 ... 0.00 0.3 \n", "1 186 43 ... 0.00 0.3 \n", "2 262 236 ... 0.00 0.3 \n", "3 188 97 ... 0.00 0.3 \n", "4 4 112 ... 0.00 0.3 \n", "5 161 236 ... 0.00 0.3 \n", "6 79 241 ... 0.00 0.3 \n", "7 237 114 ... 0.00 0.3 \n", "8 234 249 ... 0.00 0.3 \n", "9 239 237 ... 0.00 0.3 \n", "10 162 142 ... 0.00 0.3 \n", "11 236 132 ... 5.54 0.3 \n", "12 13 148 ... 0.00 0.3 \n", "13 209 25 ... 0.00 0.3 \n", "14 238 161 ... 0.00 0.3 \n", "15 239 231 ... 0.00 0.3 \n", "16 163 41 ... 0.00 0.3 \n", "17 142 100 ... 0.00 0.3 \n", "18 263 262 ... 0.00 0.3 \n", "19 79 68 ... 0.00 0.3 \n", "\n", " total_amount duration pickup_dropoff mean_distance mean_duration \\\n", "0 16.56 14.066667 100 231 3.521667 22.847222 \n", "1 20.80 26.500000 186 43 3.108889 24.470370 \n", "2 8.75 7.200000 262 236 0.881429 7.250000 \n", "3 27.69 30.250000 188 97 3.700000 30.250000 \n", "4 17.80 16.716667 4 112 4.435000 14.616667 \n", "5 12.36 8.000000 161 236 2.052258 11.855376 \n", "6 59.16 59.633333 79 241 12.830000 59.633333 \n", "7 19.58 21.983333 237 114 4.022500 26.437500 \n", "8 9.80 12.116667 234 249 1.019259 7.873457 \n", "9 16.55 20.433333 239 237 1.580000 10.541111 \n", "10 14.76 15.200000 162 142 1.641000 14.178333 \n", "11 72.92 36.800000 236 132 19.211667 40.500000 \n", "12 20.15 19.750000 13 148 3.307500 15.058333 \n", "13 12.96 7.866667 209 25 2.390000 7.866667 \n", "14 22.85 24.316667 238 161 2.930000 19.555000 \n", "15 22.80 16.133333 239 231 5.950000 17.166667 \n", "16 17.15 12.483333 163 41 3.266667 14.733333 \n", "17 10.30 12.200000 142 100 1.622857 12.304762 \n", "18 6.80 4.683333 263 262 0.662143 4.577381 \n", "19 16.00 13.066667 79 68 2.138333 13.638889 \n", "\n", " day month rush_hour \n", "0 Saturday March 0 \n", "1 Tuesday April 0 \n", "2 Friday December 1 \n", "3 Sunday May 0 \n", "4 Saturday April 0 \n", "5 Saturday March 0 \n", "6 Wednesday May 1 \n", "7 Tuesday August 1 \n", "8 Saturday February 0 \n", "9 Friday November 0 \n", "10 Saturday March 0 \n", "11 Sunday March 0 \n", "12 Friday June 1 \n", "13 Monday November 0 \n", "14 Wednesday February 0 \n", "15 Friday June 1 \n", "16 Tuesday August 1 \n", "17 Monday July 0 \n", "18 Monday April 1 \n", "19 Sunday March 0 \n", "\n", "[20 rows x 25 columns]" ] }, "execution_count": 31, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Create 'rush_hour' col\n", "### YOUR CODE HERE ###\n", "df0['rush_hour'] = 0\n", "# If day is Saturday or Sunday, impute 0 in `rush_hour` column\n", "### YOUR CODE HERE ###\n", "df0['rush_hour'][((df0['tpep_pickup_datetime'].dt.hour >= 6) & (df0['tpep_pickup_datetime'].dt.hour <= 10)) |\n", " ((df0['tpep_pickup_datetime'].dt.hour >= 16) & (df0['tpep_pickup_datetime'].dt.hour <= 20))] = 1\n", "df0['rush_hour'][(df0['day']=='Saturday') | (df0['day']=='Sunday')] = 0\n", "\n", "df0.head(20)" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [], "source": [ "### YOUR CODE HERE ###" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [], "source": [ "# Apply the `rush_hourizer()` function to the new column\n", "### YOUR CODE HERE ###" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Task 4. Scatter plot\n", "\n", "Create a scatterplot to visualize the relationship between `mean_duration` and `fare_amount`." ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Create a scatterplot to visualize the relationship between variables of interest\n", "### YOUR CODE HERE ###\n", "sns.scatterplot(data=df0, x='mean_duration', y='fare_amount')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The `mean_duration` variable correlates with the target variable. But what are the horizontal lines around fare amounts of 52 dollars and 63 dollars? What are the values and how many are there?\n", "\n", "You know what one of the lines represents. 62 dollars and 50 cents is the maximum that was imputed for outliers, so all former outliers will now have fare amounts of \\$62.50. What is the other line?\n", "\n", "Check the value of the rides in the second horizontal line in the scatter plot." ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Unnamed: 0VendorIDtpep_pickup_datetimetpep_dropoff_datetimepassenger_counttrip_distanceRatecodeIDstore_and_fwd_flagPULocationIDDOLocationID...tolls_amountimprovement_surchargetotal_amountdurationpickup_dropoffmean_distancemean_durationdaymonthrush_hour
111860005922017-03-05 19:15:302017-03-05 19:52:18218.902N236132...5.540.372.9236.800000236 13219.21166740.500000SundayMarch0
1104795979512017-06-03 14:24:572017-06-03 15:31:48118.002N132163...0.000.352.8066.850000132 16319.22900052.941667SaturdayJune0
15610488110112017-12-11 10:21:182017-12-11 11:14:57115.601N13888...5.760.369.6653.650000138 8815.39333364.316667MondayDecember1
1619572920422017-11-11 20:16:162017-11-11 20:17:1410.232N132132...0.000.352.800.966667132 1322.2558623.021839SaturdayNovember0
24710340486822017-12-06 23:37:082017-12-07 00:06:19118.932N13279...0.000.352.8029.183333132 7919.43166747.275000WednesdayDecember0
35610845874922017-12-21 21:31:122017-12-21 22:11:58618.171N132145...0.000.352.8040.766667132 14515.83714340.759524ThursdayDecember0
3798047943222017-09-24 23:45:452017-09-25 00:15:14117.992N132234...5.760.373.2029.483333132 23417.65400049.833333SundaySeptember0
3881622615712017-02-28 18:30:052017-02-28 19:09:55118.402N13248...5.540.362.8439.833333132 4818.76190558.246032TuesdayFebruary1
4065525344222017-06-05 12:51:582017-06-05 13:07:3514.732N22888...5.760.358.5615.616667228 884.73000015.616667MondayJune0
4496590002922017-08-03 22:47:142017-08-03 23:32:41218.212N13248...5.760.358.5645.450000132 4818.76190558.246032ThursdayAugust0
4688090424022017-09-26 13:48:262017-09-26 14:31:17117.272N186132...5.760.358.5642.850000186 13217.09600042.920000TuesdaySeptember0
5203370621422017-04-23 21:34:482017-04-23 22:46:23618.342N132148...0.000.357.8071.583333132 14817.99428646.340476SundayApril0
5699925987222017-11-22 21:31:322017-11-22 22:00:25118.652N132144...0.000.363.3628.883333132 14418.53750037.000000WednesdayNovember0
5726105041822017-07-18 13:29:062017-07-18 13:29:1910.002N230161...5.760.370.270.216667230 1610.6854847.965591TuesdayJuly0
5865444464722017-06-26 13:39:122017-06-26 14:34:54117.762N211132...5.760.370.2755.700000211 13216.58000061.691667MondayJune0
6929442428922017-11-07 22:15:002017-11-07 22:45:32216.972N132170...5.760.370.2730.533333132 17017.20300037.113333TuesdayNovember0
71710309422012017-12-06 05:19:502017-12-06 05:53:52120.802N132239...5.760.364.4134.033333132 23920.90125044.862500WednesdayDecember0
7196611583412017-08-04 17:53:342017-08-04 18:50:56121.602N264264...5.760.375.6657.366667264 2643.19151615.618773FridayAugust1
7825593413722017-06-09 09:31:252017-06-09 10:24:10218.812N163132...0.000.366.0052.750000163 13217.27583352.338889FridayJune1
8161373192622017-02-21 06:11:032017-02-21 06:59:39516.942N132170...5.540.360.3448.600000132 17017.20300037.113333TuesdayFebruary1
8185227774322017-06-20 08:15:182017-06-20 10:24:37117.772N132246...5.760.370.2788.783333132 24618.51500066.316667TuesdayJune1
835268430522017-01-10 22:29:472017-01-10 23:06:46118.572N13248...0.000.366.0036.983333132 4818.76190558.246032TuesdayJanuary0
8409086081422017-10-27 21:50:002017-10-27 22:35:04122.432N132163...5.760.358.5645.066667132 16319.22900052.941667FridayOctober0
86110657518612017-12-16 06:39:592017-12-16 07:07:59217.802N75132...5.760.364.5628.00000075 13218.44250036.204167SaturdayDecember0
88111049561122017-12-30 05:25:292017-12-30 06:01:29618.232N68132...0.000.352.8036.00000068 13218.78500058.041667SaturdayDecember0
9588701750312017-10-15 22:39:122017-10-15 23:14:22121.802N132261...0.000.352.8035.166667132 26122.11500051.493750SundayOctober0
9701276260822017-02-17 20:39:422017-02-17 21:13:29119.572N132140...5.540.370.0133.783333132 14019.29333336.791667FridayFebruary1
9847126444212017-08-23 18:23:262017-08-23 19:18:29116.702N132230...0.000.399.5955.050000132 23018.57120059.598000WednesdayAugust1
10821100630022017-02-07 17:20:192017-02-07 17:34:4111.092N17048...5.540.362.8414.366667170 481.26578914.135965TuesdayFebruary1
10976888203622017-08-14 23:01:152017-08-14 23:03:3552.122N265265...0.000.352.802.333333265 2650.7530773.411538MondayAugust0
\n", "

30 rows × 25 columns

\n", "
" ], "text/plain": [ " Unnamed: 0 VendorID tpep_pickup_datetime tpep_dropoff_datetime \\\n", "11 18600059 2 2017-03-05 19:15:30 2017-03-05 19:52:18 \n", "110 47959795 1 2017-06-03 14:24:57 2017-06-03 15:31:48 \n", "156 104881101 1 2017-12-11 10:21:18 2017-12-11 11:14:57 \n", "161 95729204 2 2017-11-11 20:16:16 2017-11-11 20:17:14 \n", "247 103404868 2 2017-12-06 23:37:08 2017-12-07 00:06:19 \n", "356 108458749 2 2017-12-21 21:31:12 2017-12-21 22:11:58 \n", "379 80479432 2 2017-09-24 23:45:45 2017-09-25 00:15:14 \n", "388 16226157 1 2017-02-28 18:30:05 2017-02-28 19:09:55 \n", "406 55253442 2 2017-06-05 12:51:58 2017-06-05 13:07:35 \n", "449 65900029 2 2017-08-03 22:47:14 2017-08-03 23:32:41 \n", "468 80904240 2 2017-09-26 13:48:26 2017-09-26 14:31:17 \n", "520 33706214 2 2017-04-23 21:34:48 2017-04-23 22:46:23 \n", "569 99259872 2 2017-11-22 21:31:32 2017-11-22 22:00:25 \n", "572 61050418 2 2017-07-18 13:29:06 2017-07-18 13:29:19 \n", "586 54444647 2 2017-06-26 13:39:12 2017-06-26 14:34:54 \n", "692 94424289 2 2017-11-07 22:15:00 2017-11-07 22:45:32 \n", "717 103094220 1 2017-12-06 05:19:50 2017-12-06 05:53:52 \n", "719 66115834 1 2017-08-04 17:53:34 2017-08-04 18:50:56 \n", "782 55934137 2 2017-06-09 09:31:25 2017-06-09 10:24:10 \n", "816 13731926 2 2017-02-21 06:11:03 2017-02-21 06:59:39 \n", "818 52277743 2 2017-06-20 08:15:18 2017-06-20 10:24:37 \n", "835 2684305 2 2017-01-10 22:29:47 2017-01-10 23:06:46 \n", "840 90860814 2 2017-10-27 21:50:00 2017-10-27 22:35:04 \n", "861 106575186 1 2017-12-16 06:39:59 2017-12-16 07:07:59 \n", "881 110495611 2 2017-12-30 05:25:29 2017-12-30 06:01:29 \n", "958 87017503 1 2017-10-15 22:39:12 2017-10-15 23:14:22 \n", "970 12762608 2 2017-02-17 20:39:42 2017-02-17 21:13:29 \n", "984 71264442 1 2017-08-23 18:23:26 2017-08-23 19:18:29 \n", "1082 11006300 2 2017-02-07 17:20:19 2017-02-07 17:34:41 \n", "1097 68882036 2 2017-08-14 23:01:15 2017-08-14 23:03:35 \n", "\n", " passenger_count trip_distance RatecodeID store_and_fwd_flag \\\n", "11 2 18.90 2 N \n", "110 1 18.00 2 N \n", "156 1 15.60 1 N \n", "161 1 0.23 2 N \n", "247 1 18.93 2 N \n", "356 6 18.17 1 N \n", "379 1 17.99 2 N \n", "388 1 18.40 2 N \n", "406 1 4.73 2 N \n", "449 2 18.21 2 N \n", "468 1 17.27 2 N \n", "520 6 18.34 2 N \n", "569 1 18.65 2 N \n", "572 1 0.00 2 N \n", "586 1 17.76 2 N \n", "692 2 16.97 2 N \n", "717 1 20.80 2 N \n", "719 1 21.60 2 N \n", "782 2 18.81 2 N \n", "816 5 16.94 2 N \n", "818 1 17.77 2 N \n", "835 1 18.57 2 N \n", "840 1 22.43 2 N \n", "861 2 17.80 2 N \n", "881 6 18.23 2 N \n", "958 1 21.80 2 N \n", "970 1 19.57 2 N \n", "984 1 16.70 2 N \n", "1082 1 1.09 2 N \n", "1097 5 2.12 2 N \n", "\n", " PULocationID DOLocationID ... tolls_amount improvement_surcharge \\\n", "11 236 132 ... 5.54 0.3 \n", "110 132 163 ... 0.00 0.3 \n", "156 138 88 ... 5.76 0.3 \n", "161 132 132 ... 0.00 0.3 \n", "247 132 79 ... 0.00 0.3 \n", "356 132 145 ... 0.00 0.3 \n", "379 132 234 ... 5.76 0.3 \n", "388 132 48 ... 5.54 0.3 \n", "406 228 88 ... 5.76 0.3 \n", "449 132 48 ... 5.76 0.3 \n", "468 186 132 ... 5.76 0.3 \n", "520 132 148 ... 0.00 0.3 \n", "569 132 144 ... 0.00 0.3 \n", "572 230 161 ... 5.76 0.3 \n", "586 211 132 ... 5.76 0.3 \n", "692 132 170 ... 5.76 0.3 \n", "717 132 239 ... 5.76 0.3 \n", "719 264 264 ... 5.76 0.3 \n", "782 163 132 ... 0.00 0.3 \n", "816 132 170 ... 5.54 0.3 \n", "818 132 246 ... 5.76 0.3 \n", "835 132 48 ... 0.00 0.3 \n", "840 132 163 ... 5.76 0.3 \n", "861 75 132 ... 5.76 0.3 \n", "881 68 132 ... 0.00 0.3 \n", "958 132 261 ... 0.00 0.3 \n", "970 132 140 ... 5.54 0.3 \n", "984 132 230 ... 0.00 0.3 \n", "1082 170 48 ... 5.54 0.3 \n", "1097 265 265 ... 0.00 0.3 \n", "\n", " total_amount duration pickup_dropoff mean_distance mean_duration \\\n", "11 72.92 36.800000 236 132 19.211667 40.500000 \n", "110 52.80 66.850000 132 163 19.229000 52.941667 \n", "156 69.66 53.650000 138 88 15.393333 64.316667 \n", "161 52.80 0.966667 132 132 2.255862 3.021839 \n", "247 52.80 29.183333 132 79 19.431667 47.275000 \n", "356 52.80 40.766667 132 145 15.837143 40.759524 \n", "379 73.20 29.483333 132 234 17.654000 49.833333 \n", "388 62.84 39.833333 132 48 18.761905 58.246032 \n", "406 58.56 15.616667 228 88 4.730000 15.616667 \n", "449 58.56 45.450000 132 48 18.761905 58.246032 \n", "468 58.56 42.850000 186 132 17.096000 42.920000 \n", "520 57.80 71.583333 132 148 17.994286 46.340476 \n", "569 63.36 28.883333 132 144 18.537500 37.000000 \n", "572 70.27 0.216667 230 161 0.685484 7.965591 \n", "586 70.27 55.700000 211 132 16.580000 61.691667 \n", "692 70.27 30.533333 132 170 17.203000 37.113333 \n", "717 64.41 34.033333 132 239 20.901250 44.862500 \n", "719 75.66 57.366667 264 264 3.191516 15.618773 \n", "782 66.00 52.750000 163 132 17.275833 52.338889 \n", "816 60.34 48.600000 132 170 17.203000 37.113333 \n", "818 70.27 88.783333 132 246 18.515000 66.316667 \n", "835 66.00 36.983333 132 48 18.761905 58.246032 \n", "840 58.56 45.066667 132 163 19.229000 52.941667 \n", "861 64.56 28.000000 75 132 18.442500 36.204167 \n", "881 52.80 36.000000 68 132 18.785000 58.041667 \n", "958 52.80 35.166667 132 261 22.115000 51.493750 \n", "970 70.01 33.783333 132 140 19.293333 36.791667 \n", "984 99.59 55.050000 132 230 18.571200 59.598000 \n", "1082 62.84 14.366667 170 48 1.265789 14.135965 \n", "1097 52.80 2.333333 265 265 0.753077 3.411538 \n", "\n", " day month rush_hour \n", "11 Sunday March 0 \n", "110 Saturday June 0 \n", "156 Monday December 1 \n", "161 Saturday November 0 \n", "247 Wednesday December 0 \n", "356 Thursday December 0 \n", "379 Sunday September 0 \n", "388 Tuesday February 1 \n", "406 Monday June 0 \n", "449 Thursday August 0 \n", "468 Tuesday September 0 \n", "520 Sunday April 0 \n", "569 Wednesday November 0 \n", "572 Tuesday July 0 \n", "586 Monday June 0 \n", "692 Tuesday November 0 \n", "717 Wednesday December 0 \n", "719 Friday August 1 \n", "782 Friday June 1 \n", "816 Tuesday February 1 \n", "818 Tuesday June 1 \n", "835 Tuesday January 0 \n", "840 Friday October 0 \n", "861 Saturday December 0 \n", "881 Saturday December 0 \n", "958 Sunday October 0 \n", "970 Friday February 1 \n", "984 Wednesday August 1 \n", "1082 Tuesday February 1 \n", "1097 Monday August 0 \n", "\n", "[30 rows x 25 columns]" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" } ], "source": [ "### YOUR CODE HERE ###\n", "df0[(df0['fare_amount']>51) & (df0['fare_amount']<53)].head(30)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Examine the first 30 of these trips." ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [], "source": [ "# Set pandas to display all columns\n", "### YOUR CODE HERE ###" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Question:** What do you notice about the first 30 trips?\n", "\n", "==> ENTER YOUR RESPONSE HERE\n", "I noticed all of them have the same fare of 52, that seems to be a default amount for some unknonw reason." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Task 5. Isolate modeling variables\n", "\n", "Drop features that are redundant, irrelevant, or that will not be available in a deployed environment." ] }, { "cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Int64Index: 22699 entries, 0 to 22698\n", "Data columns (total 25 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 Unnamed: 0 22699 non-null int64 \n", " 1 VendorID 22699 non-null int64 \n", " 2 tpep_pickup_datetime 22699 non-null datetime64[ns]\n", " 3 tpep_dropoff_datetime 22699 non-null datetime64[ns]\n", " 4 passenger_count 22699 non-null int64 \n", " 5 trip_distance 22699 non-null float64 \n", " 6 RatecodeID 22699 non-null int64 \n", " 7 store_and_fwd_flag 22699 non-null object \n", " 8 PULocationID 22699 non-null int64 \n", " 9 DOLocationID 22699 non-null int64 \n", " 10 payment_type 22699 non-null int64 \n", " 11 fare_amount 22699 non-null float64 \n", " 12 extra 22699 non-null float64 \n", " 13 mta_tax 22699 non-null float64 \n", " 14 tip_amount 22699 non-null float64 \n", " 15 tolls_amount 22699 non-null float64 \n", " 16 improvement_surcharge 22699 non-null float64 \n", " 17 total_amount 22699 non-null float64 \n", " 18 duration 22699 non-null float64 \n", " 19 pickup_dropoff 22699 non-null object \n", " 20 mean_distance 22699 non-null float64 \n", " 21 mean_duration 22699 non-null float64 \n", " 22 day 22699 non-null object \n", " 23 month 22699 non-null object \n", " 24 rush_hour 22699 non-null int64 \n", "dtypes: datetime64[ns](2), float64(11), int64(8), object(4)\n", "memory usage: 4.5+ MB\n" ] } ], "source": [ "### YOUR CODE HERE ###\n", "df0.info()" ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
VendorIDpassenger_countRatecodeIDpayment_typefare_amounttolls_amountmean_distancemean_durationrush_hour
0261113.00.03.52166722.8472220
1111116.00.03.10888924.4703700
211116.50.00.8814297.2500001
3211120.50.03.70000030.2500000
4211216.50.04.43500014.6166670
526119.00.02.05225811.8553760
6211147.50.012.83000059.6333331
7211116.00.04.02250026.4375001
821129.00.01.0192597.8734570
9111113.00.01.58000010.5411110
\n", "
" ], "text/plain": [ " VendorID passenger_count RatecodeID payment_type fare_amount \\\n", "0 2 6 1 1 13.0 \n", "1 1 1 1 1 16.0 \n", "2 1 1 1 1 6.5 \n", "3 2 1 1 1 20.5 \n", "4 2 1 1 2 16.5 \n", "5 2 6 1 1 9.0 \n", "6 2 1 1 1 47.5 \n", "7 2 1 1 1 16.0 \n", "8 2 1 1 2 9.0 \n", "9 1 1 1 1 13.0 \n", "\n", " tolls_amount mean_distance mean_duration rush_hour \n", "0 0.0 3.521667 22.847222 0 \n", "1 0.0 3.108889 24.470370 0 \n", "2 0.0 0.881429 7.250000 1 \n", "3 0.0 3.700000 30.250000 0 \n", "4 0.0 4.435000 14.616667 0 \n", "5 0.0 2.052258 11.855376 0 \n", "6 0.0 12.830000 59.633333 1 \n", "7 0.0 4.022500 26.437500 1 \n", "8 0.0 1.019259 7.873457 0 \n", "9 0.0 1.580000 10.541111 0 " ] }, "execution_count": 38, "metadata": {}, "output_type": "execute_result" } ], "source": [ "### YOUR CODE HERE ###\n", "df = df0.drop(columns=[\"Unnamed: 0\", \"tpep_pickup_datetime\", \"tpep_dropoff_datetime\",\n", " \"trip_distance\", \"store_and_fwd_flag\", \"PULocationID\", \"DOLocationID\",\n", " \"extra\", \"mta_tax\", \"tip_amount\", \"improvement_surcharge\", \"total_amount\",\n", " \"duration\", \"pickup_dropoff\", \"day\", \"month\"])\n", "df.head(10)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Task 6. Pair plot\n", "\n", "Create a pairplot to visualize pairwise relationships between `fare_amount`, `mean_duration`, and `mean_distance`." ] }, { "cell_type": "code", "execution_count": 39, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 39, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Create a pairplot to visualize pairwise relationships between variables in the data\n", "### YOUR CODE HERE ###\n", "sns.pairplot(df)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "These variables all show linear correlation with each other. Investigate this further." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Task 7. Identify correlations" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Next, code a correlation matrix to help determine most correlated variables." ] }, { "cell_type": "code", "execution_count": 40, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
VendorIDpassenger_countRatecodeIDpayment_typefare_amounttolls_amountmean_distancemean_durationrush_hour
VendorID1.0000000.266463-0.002991-0.0177870.0010450.0111220.0047410.001876-0.000752
passenger_count0.2664631.000000-0.0057430.0161780.0149420.0095320.0134280.015852-0.024283
RatecodeID-0.002991-0.0057431.000000-0.0009820.2221020.1758600.1593530.1116670.004145
payment_type-0.0177870.016178-0.0009821.000000-0.049516-0.041217-0.044495-0.054298-0.049030
fare_amount0.0010450.0149420.222102-0.0495161.0000000.6167190.9101850.859105-0.025901
tolls_amount0.0111220.0095320.175860-0.0412170.6167191.0000000.6212290.512261-0.000694
mean_distance0.0047410.0134280.159353-0.0444950.9101850.6212291.0000000.874864-0.046794
mean_duration0.0018760.0158520.111667-0.0542980.8591050.5122610.8748641.000000-0.027499
rush_hour-0.000752-0.0242830.004145-0.049030-0.025901-0.000694-0.046794-0.0274991.000000
\n", "
" ], "text/plain": [ " VendorID passenger_count RatecodeID payment_type \\\n", "VendorID 1.000000 0.266463 -0.002991 -0.017787 \n", "passenger_count 0.266463 1.000000 -0.005743 0.016178 \n", "RatecodeID -0.002991 -0.005743 1.000000 -0.000982 \n", "payment_type -0.017787 0.016178 -0.000982 1.000000 \n", "fare_amount 0.001045 0.014942 0.222102 -0.049516 \n", "tolls_amount 0.011122 0.009532 0.175860 -0.041217 \n", "mean_distance 0.004741 0.013428 0.159353 -0.044495 \n", "mean_duration 0.001876 0.015852 0.111667 -0.054298 \n", "rush_hour -0.000752 -0.024283 0.004145 -0.049030 \n", "\n", " fare_amount tolls_amount mean_distance mean_duration \\\n", "VendorID 0.001045 0.011122 0.004741 0.001876 \n", "passenger_count 0.014942 0.009532 0.013428 0.015852 \n", "RatecodeID 0.222102 0.175860 0.159353 0.111667 \n", "payment_type -0.049516 -0.041217 -0.044495 -0.054298 \n", "fare_amount 1.000000 0.616719 0.910185 0.859105 \n", "tolls_amount 0.616719 1.000000 0.621229 0.512261 \n", "mean_distance 0.910185 0.621229 1.000000 0.874864 \n", "mean_duration 0.859105 0.512261 0.874864 1.000000 \n", "rush_hour -0.025901 -0.000694 -0.046794 -0.027499 \n", "\n", " rush_hour \n", "VendorID -0.000752 \n", "passenger_count -0.024283 \n", "RatecodeID 0.004145 \n", "payment_type -0.049030 \n", "fare_amount -0.025901 \n", "tolls_amount -0.000694 \n", "mean_distance -0.046794 \n", "mean_duration -0.027499 \n", "rush_hour 1.000000 " ] }, "execution_count": 40, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Correlation matrix to help determine most correlated variables\n", "### YOUR CODE HERE ###\n", "corr_matrix = df.corr()\n", "corr_matrix" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Visualize a correlation heatmap of the data." ] }, { "cell_type": "code", "execution_count": 41, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 41, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Create correlation heatmap\n", "### YOUR CODE HERE ###\n", "sns.heatmap(corr_matrix)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Question:** Which variable(s) are correlated with the target variable of `fare_amount`? \n", "\n", "Try modeling with both variables even though they are correlated." ] }, { "cell_type": "markdown", "metadata": { "id": "lgPul2DiY6T4" }, "source": [ "\n", "\n", "## PACE: **Construct**\n", "\n", "After analysis and deriving variables with close relationships, it is time to begin constructing the model. Consider the questions in your PACE Strategy Document to reflect on the Construct stage.\n" ] }, { "cell_type": "markdown", "metadata": { "id": "P_QYzJfVUrIc" }, "source": [ "### Task 8a. Split data into outcome variable and features" ] }, { "cell_type": "code", "execution_count": 42, "metadata": { "id": "AzcDgLRET4d7" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Int64Index: 22699 entries, 0 to 22698\n", "Data columns (total 1 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 fare_amount 22699 non-null float64\n", "dtypes: float64(1)\n", "memory usage: 870.7 KB\n" ] } ], "source": [ "### YOUR CODE HERE ###\n", "y = df[['fare_amount']]\n", "X = df.drop(columns='fare_amount')\n", "y.info()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Set your X and y variables. X represents the features and y represents the outcome (target) variable." ] }, { "cell_type": "code", "execution_count": 43, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Int64Index: 22699 entries, 0 to 22698\n", "Data columns (total 8 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 VendorID 22699 non-null int64 \n", " 1 passenger_count 22699 non-null int64 \n", " 2 RatecodeID 22699 non-null int64 \n", " 3 payment_type 22699 non-null int64 \n", " 4 tolls_amount 22699 non-null float64\n", " 5 mean_distance 22699 non-null float64\n", " 6 mean_duration 22699 non-null float64\n", " 7 rush_hour 22699 non-null int64 \n", "dtypes: float64(3), int64(5)\n", "memory usage: 2.1 MB\n" ] } ], "source": [ "# Remove the target column from the features\n", "# X = df2.drop(columns='fare_amount')\n", "### YOUR CODE HERE ###\n", "# Set y variable\n", "### YOUR CODE HERE ###\n", "# Display first few rows\n", "### YOUR CODE HERE ###\n", "X.info()" ] }, { "cell_type": "markdown", "metadata": { "id": "o3ArC_5xa7Oi" }, "source": [ "### Task 8b. Pre-process data\n" ] }, { "cell_type": "markdown", "metadata": { "id": "PdfTaopCcbTj" }, "source": [ "Dummy encode categorical variables" ] }, { "cell_type": "code", "execution_count": 44, "metadata": { "id": "4Y3T2poF28fP" }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
passenger_countRatecodeIDpayment_typetolls_amountmean_distancemean_durationrush_hourVendorID_1VendorID_2
06110.003.52166722.847222001
11110.003.10888924.470370010
21110.000.8814297.250000110
31110.003.70000030.250000001
41120.004.43500014.616667001
..............................
226943120.001.0982148.594643101
226951215.7618.75750059.560417001
226961120.000.6842426.609091001
226971110.002.07750016.650000001
226981110.001.4769709.405556010
\n", "

22699 rows × 9 columns

\n", "
" ], "text/plain": [ " passenger_count RatecodeID payment_type tolls_amount mean_distance \\\n", "0 6 1 1 0.00 3.521667 \n", "1 1 1 1 0.00 3.108889 \n", "2 1 1 1 0.00 0.881429 \n", "3 1 1 1 0.00 3.700000 \n", "4 1 1 2 0.00 4.435000 \n", "... ... ... ... ... ... \n", "22694 3 1 2 0.00 1.098214 \n", "22695 1 2 1 5.76 18.757500 \n", "22696 1 1 2 0.00 0.684242 \n", "22697 1 1 1 0.00 2.077500 \n", "22698 1 1 1 0.00 1.476970 \n", "\n", " mean_duration rush_hour VendorID_1 VendorID_2 \n", "0 22.847222 0 0 1 \n", "1 24.470370 0 1 0 \n", "2 7.250000 1 1 0 \n", "3 30.250000 0 0 1 \n", "4 14.616667 0 0 1 \n", "... ... ... ... ... \n", "22694 8.594643 1 0 1 \n", "22695 59.560417 0 0 1 \n", "22696 6.609091 0 0 1 \n", "22697 16.650000 0 0 1 \n", "22698 9.405556 0 1 0 \n", "\n", "[22699 rows x 9 columns]" ] }, "execution_count": 44, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Convert VendorID to string\n", "### YOUR CODE HERE ###\n", "X['VendorID'] = X['VendorID'].astype(str)\n", "# Get dummies\n", "### YOUR CODE HERE ###\n", "Xd = pd.get_dummies(X, columns=['VendorID'])\n", "Xd" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Split data into training and test sets" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Create training and testing sets. The test set should contain 20% of the total samples. Set `random_state=0`." ] }, { "cell_type": "code", "execution_count": 139, "metadata": { "id": "A2BNUvacwaZY" }, "outputs": [], "source": [ "# Create training and testing sets\n", "#### YOUR CODE HERE ####\n", "X_train, X_test, y_train, y_test = train_test_split(X,y, test_size=0.3, random_state=42)" ] }, { "cell_type": "markdown", "metadata": { "id": "iDYyjWssbnBG" }, "source": [ "### Standardize the data\n", "\n", "Use `StandardScaler()`, `fit()`, and `transform()` to standardize the `X_train` variables. Assign the results to a variable called `X_train_scaled`." ] }, { "cell_type": "code", "execution_count": 46, "metadata": {}, "outputs": [], "source": [ "# Standardize the X variables\n", "### YOUR CODE HERE ###\n", "from sklearn.preprocessing import StandardScaler\n", "\n", "scaler = StandardScaler()\n", "\n", "# 1. Fit & transform training set in one step\n", "X_train_scaled = scaler.fit_transform(X_train)\n", "\n", "# 2. ONLY transform test set (uses training mean/std)\n", "X_test_scaled = scaler.transform(X_test)" ] }, { "cell_type": "code", "execution_count": 98, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
VendorIDpassenger_countRatecodeIDpayment_typetolls_amountmean_distancemean_durationrush_hour
0-1.116436-0.497569-0.0561361.329293-0.225058-0.407526-0.5162401.281921
10.895707-0.497569-0.056136-0.681095-0.225058-0.513546-0.514487-0.780079
20.895707-0.497569-0.056136-0.681095-0.225058-0.3788060.0318861.281921
30.895707-0.497569-0.056136-0.681095-0.225058-0.436989-0.709230-0.780079
40.895707-0.497569-0.0561361.329293-0.2250580.0955660.048359-0.780079
50.895707-0.497569-0.0561361.329293-0.225058-0.515608-0.6322291.281921
6-1.116436-0.497569-0.056136-0.681095-0.225058-0.551298-0.684252-0.780079
7-1.1164360.286433-0.0561361.329293-0.225058-0.0891310.0303891.281921
80.8957072.638436-0.056136-0.681095-0.2250580.3722840.240797-0.780079
9-1.116436-0.497569-0.056136-0.681095-0.225058-0.401395-0.1018611.281921
100.895707-0.497569-0.056136-0.681095-0.2250580.0722610.326170-0.780079
110.8957071.070434-0.056136-0.681095-0.225058-0.539575-0.814177-0.780079
12-1.116436-0.497569-0.0561365.350069-0.225058-0.616428-0.7591161.281921
13-1.1164360.286433-0.0561361.329293-0.2250580.2534120.8574671.281921
14-1.116436-0.497569-0.056136-0.681095-0.2250580.0714900.102894-0.780079
150.895707-0.497569-0.056136-0.681095-0.225058-0.364229-0.255531-0.780079
16-1.116436-0.497569-0.056136-0.681095-0.225058-0.616074-0.907399-0.780079
170.895707-0.497569-0.056136-0.681095-0.225058-0.676734-1.037558-0.780079
180.895707-0.497569-0.056136-0.681095-0.2250580.0237420.273280-0.780079
190.8957070.286433-0.0561361.329293-0.225058-0.351524-0.196280-0.780079
20-1.116436-0.497569-0.056136-0.681095-0.225058-0.315891-0.693685-0.780079
210.895707-0.497569-0.056136-0.681095-0.2250580.3633750.7894211.281921
220.8957073.422437-0.056136-0.681095-0.225058-0.2704210.274398-0.780079
230.895707-0.497569-0.056136-0.681095-0.2250580.3605921.177789-0.780079
240.8957070.286433-0.0561361.329293-0.225058-0.584674-0.749212-0.780079
250.895707-0.497569-0.056136-0.681095-0.225058-0.538203-0.6421001.281921
26-1.116436-0.497569-0.056136-0.681095-0.225058-0.230856-0.1557121.281921
27-1.116436-0.497569-0.056136-0.681095-0.225058-0.113595-0.1045881.281921
28-1.116436-0.497569-0.056136-0.681095-0.2250583.5843212.245118-0.780079
29-1.1164360.286433-0.056136-0.681095-0.2250580.6046450.436509-0.780079
\n", "
" ], "text/plain": [ " VendorID passenger_count RatecodeID payment_type tolls_amount \\\n", "0 -1.116436 -0.497569 -0.056136 1.329293 -0.225058 \n", "1 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "2 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "3 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "4 0.895707 -0.497569 -0.056136 1.329293 -0.225058 \n", "5 0.895707 -0.497569 -0.056136 1.329293 -0.225058 \n", "6 -1.116436 -0.497569 -0.056136 -0.681095 -0.225058 \n", "7 -1.116436 0.286433 -0.056136 1.329293 -0.225058 \n", "8 0.895707 2.638436 -0.056136 -0.681095 -0.225058 \n", "9 -1.116436 -0.497569 -0.056136 -0.681095 -0.225058 \n", "10 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "11 0.895707 1.070434 -0.056136 -0.681095 -0.225058 \n", "12 -1.116436 -0.497569 -0.056136 5.350069 -0.225058 \n", "13 -1.116436 0.286433 -0.056136 1.329293 -0.225058 \n", "14 -1.116436 -0.497569 -0.056136 -0.681095 -0.225058 \n", "15 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "16 -1.116436 -0.497569 -0.056136 -0.681095 -0.225058 \n", "17 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "18 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "19 0.895707 0.286433 -0.056136 1.329293 -0.225058 \n", "20 -1.116436 -0.497569 -0.056136 -0.681095 -0.225058 \n", "21 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "22 0.895707 3.422437 -0.056136 -0.681095 -0.225058 \n", "23 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "24 0.895707 0.286433 -0.056136 1.329293 -0.225058 \n", "25 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "26 -1.116436 -0.497569 -0.056136 -0.681095 -0.225058 \n", "27 -1.116436 -0.497569 -0.056136 -0.681095 -0.225058 \n", "28 -1.116436 -0.497569 -0.056136 -0.681095 -0.225058 \n", "29 -1.116436 0.286433 -0.056136 -0.681095 -0.225058 \n", "\n", " mean_distance mean_duration rush_hour \n", "0 -0.407526 -0.516240 1.281921 \n", "1 -0.513546 -0.514487 -0.780079 \n", "2 -0.378806 0.031886 1.281921 \n", "3 -0.436989 -0.709230 -0.780079 \n", "4 0.095566 0.048359 -0.780079 \n", "5 -0.515608 -0.632229 1.281921 \n", "6 -0.551298 -0.684252 -0.780079 \n", "7 -0.089131 0.030389 1.281921 \n", "8 0.372284 0.240797 -0.780079 \n", "9 -0.401395 -0.101861 1.281921 \n", "10 0.072261 0.326170 -0.780079 \n", "11 -0.539575 -0.814177 -0.780079 \n", "12 -0.616428 -0.759116 1.281921 \n", "13 0.253412 0.857467 1.281921 \n", "14 0.071490 0.102894 -0.780079 \n", "15 -0.364229 -0.255531 -0.780079 \n", "16 -0.616074 -0.907399 -0.780079 \n", "17 -0.676734 -1.037558 -0.780079 \n", "18 0.023742 0.273280 -0.780079 \n", "19 -0.351524 -0.196280 -0.780079 \n", "20 -0.315891 -0.693685 -0.780079 \n", "21 0.363375 0.789421 1.281921 \n", "22 -0.270421 0.274398 -0.780079 \n", "23 0.360592 1.177789 -0.780079 \n", "24 -0.584674 -0.749212 -0.780079 \n", "25 -0.538203 -0.642100 1.281921 \n", "26 -0.230856 -0.155712 1.281921 \n", "27 -0.113595 -0.104588 1.281921 \n", "28 3.584321 2.245118 -0.780079 \n", "29 0.604645 0.436509 -0.780079 " ] }, "execution_count": 98, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X_train_scaled.head(30)" ] }, { "cell_type": "code", "execution_count": 79, "metadata": {}, "outputs": [], "source": [ "X_train_scaled = pd.DataFrame(X_train_scaled, columns=[\"VendorID\",\"passenger_count\",\"RatecodeID\",\"payment_type\",\"tolls_amount\",\n", " \"mean_distance\",\"mean_duration\",\"rush_hour\"])\n", "X_test_scaled = pd.DataFrame(X_test_scaled, columns=[\"VendorID\",\"passenger_count\",\"RatecodeID\",\"payment_type\",\"tolls_amount\",\n", " \"mean_distance\",\"mean_duration\",\"rush_hour\"])" ] }, { "cell_type": "code", "execution_count": 140, "metadata": {}, "outputs": [], "source": [ "y_train=y_train.reset_index(drop=True)\n", "y_test = y_test.reset_index(drop=True)" ] }, { "cell_type": "markdown", "metadata": { "id": "wk0rjKeO3JLv" }, "source": [ "### Fit the model\n", "\n", "Instantiate your model and fit it to the training data." ] }, { "cell_type": "code", "execution_count": 99, "metadata": { "id": "SClNm5hWotj6" }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
OLS Regression Results
Dep. Variable: y_train R-squared: 0.870
Model: OLS Adj. R-squared: 0.870
Method: Least Squares F-statistic: 1.178e+04
Date: Sun, 27 Sep 2026 Prob (F-statistic): 0.00
Time: 20:31:20 Log-Likelihood: -43896.
No. Observations: 15889 AIC: 8.781e+04
Df Residuals: 15879 BIC: 8.789e+04
Df Model: 9
Covariance Type: nonrobust
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
coef std err t P>|t| [0.025 0.975]
Intercept 12.6749 0.052 244.336 0.000 12.573 12.777
C(rush_hour)[T.1.281920660157489] 0.1835 0.063 2.918 0.004 0.060 0.307
C(RatecodeID)[T.1.1524551660464013] 7.0621 0.255 27.655 0.000 6.562 7.563
C(RatecodeID)[T.2.3610460269342153] 15.1546 0.710 21.343 0.000 13.763 16.546
C(RatecodeID)[T.3.569636887822029] 9.6482 1.464 6.590 0.000 6.779 12.518
C(RatecodeID)[T.4.778227748709843] 23.1248 0.563 41.058 0.000 22.021 24.229
C(RatecodeID)[T.118.38576867216435] 48.8707 3.835 12.743 0.000 41.354 56.388
C(VendorID)[T.0.8957071444206769] -0.0734 0.061 -1.198 0.231 -0.193 0.047
mean_distance 5.9189 0.072 82.357 0.000 5.778 6.060
mean_duration 3.3798 0.064 52.930 0.000 3.255 3.505
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
Omnibus: 12423.203 Durbin-Watson: 2.010
Prob(Omnibus): 0.000 Jarque-Bera (JB): 878888.645
Skew: 3.212 Prob(JB): 0.00
Kurtosis: 38.865 Cond. No. 173.


Warnings:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified." ], "text/plain": [ "\n", "\"\"\"\n", " OLS Regression Results \n", "==============================================================================\n", "Dep. Variable: y_train R-squared: 0.870\n", "Model: OLS Adj. R-squared: 0.870\n", "Method: Least Squares F-statistic: 1.178e+04\n", "Date: Sun, 27 Sep 2026 Prob (F-statistic): 0.00\n", "Time: 20:31:20 Log-Likelihood: -43896.\n", "No. Observations: 15889 AIC: 8.781e+04\n", "Df Residuals: 15879 BIC: 8.789e+04\n", "Df Model: 9 \n", "Covariance Type: nonrobust \n", "=======================================================================================================\n", " coef std err t P>|t| [0.025 0.975]\n", "-------------------------------------------------------------------------------------------------------\n", "Intercept 12.6749 0.052 244.336 0.000 12.573 12.777\n", "C(rush_hour)[T.1.281920660157489] 0.1835 0.063 2.918 0.004 0.060 0.307\n", "C(RatecodeID)[T.1.1524551660464013] 7.0621 0.255 27.655 0.000 6.562 7.563\n", "C(RatecodeID)[T.2.3610460269342153] 15.1546 0.710 21.343 0.000 13.763 16.546\n", "C(RatecodeID)[T.3.569636887822029] 9.6482 1.464 6.590 0.000 6.779 12.518\n", "C(RatecodeID)[T.4.778227748709843] 23.1248 0.563 41.058 0.000 22.021 24.229\n", "C(RatecodeID)[T.118.38576867216435] 48.8707 3.835 12.743 0.000 41.354 56.388\n", "C(VendorID)[T.0.8957071444206769] -0.0734 0.061 -1.198 0.231 -0.193 0.047\n", "mean_distance 5.9189 0.072 82.357 0.000 5.778 6.060\n", "mean_duration 3.3798 0.064 52.930 0.000 3.255 3.505\n", "==============================================================================\n", "Omnibus: 12423.203 Durbin-Watson: 2.010\n", "Prob(Omnibus): 0.000 Jarque-Bera (JB): 878888.645\n", "Skew: 3.212 Prob(JB): 0.00\n", "Kurtosis: 38.865 Cond. No. 173.\n", "==============================================================================\n", "\n", "Warnings:\n", "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n", "\"\"\"" ] }, "execution_count": 99, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Fit your model to the training data\n", "### YOUR CODE HERE ###\n", "ols_formula = \"y_train ~ mean_distance + mean_duration + C(rush_hour) + C(RatecodeID) + C(VendorID)\"\n", "#ols_formula = \"y_train ~ mean_distance + mean_duration + tolls_amount + C(RatecodeID)\"\n", "#ols_formula = \"y_train ~ mean_distance + mean_duration + C(rush_hour)\"\n", "#ols_formula = \"y_train ~ mean_distance + C(rush_hour)\"\n", "ols_data = pd.concat([X_train_scaled, y_train], axis=1)\n", "OLS = ols(formula=ols_formula, data=ols_data)\n", "model = OLS.fit()\n", "result = model.summary()\n", "result" ] }, { "cell_type": "markdown", "metadata": { "id": "HMj6QkK1cLmS" }, "source": [ "### Task 8c. Evaluate model" ] }, { "cell_type": "markdown", "metadata": { "id": "AromLx7t5hjt" }, "source": [ "### Train data\n", "\n", "Evaluate your model performance by calculating the residual sum of squares and the explained variance score (R^2). Calculate the Mean Absolute Error, Mean Squared Error, and the Root Mean Squared Error." ] }, { "cell_type": "code", "execution_count": 74, "metadata": { "id": "33rE1x9e3U6t" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "233481.58710770003 0.8697544929332063 2.231021348224999 14.703796656445622 3.834552993041773\n" ] } ], "source": [ "# Evaluate the model performance on the training data\n", "### YOUR CODE HERE ###\n", "resid = model.resid\n", "\n", "rss = model.ssr\n", "r2 = model.rsquared\n", "mae = np.mean(np.abs(resid))\n", "mse = model.mse_resid\n", "rmse = np.sqrt(mse)\n", "\n", "print(rss, r2, mae, mse, rmse)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Test data\n", "\n", "Calculate the same metrics on the test data. Remember to scale the `X_test` data using the scaler that was fit to the training data. Do not refit the scaler to the testing data, just transform it. Call the results `X_test_scaled`." ] }, { "cell_type": "code", "execution_count": 116, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
VendorIDpassenger_countRatecodeIDpayment_typetolls_amountmean_distancemean_durationrush_hour
00.895707-0.497569-0.056136-0.681095-0.225058-0.408106-0.111065-0.780079
10.895707-0.497569-0.056136-0.681095-0.225058-0.471788-0.3987821.281921
20.8957072.638436-0.056136-0.681095-0.225058-0.474121-0.102173-0.780079
30.8957072.638436-0.056136-0.681095-0.2250580.7934840.717148-0.780079
4-1.116436-0.497569-0.056136-0.681095-0.2250580.421837-0.216407-0.780079
\n", "
" ], "text/plain": [ " VendorID passenger_count RatecodeID payment_type tolls_amount \\\n", "0 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "1 0.895707 -0.497569 -0.056136 -0.681095 -0.225058 \n", "2 0.895707 2.638436 -0.056136 -0.681095 -0.225058 \n", "3 0.895707 2.638436 -0.056136 -0.681095 -0.225058 \n", "4 -1.116436 -0.497569 -0.056136 -0.681095 -0.225058 \n", "\n", " mean_distance mean_duration rush_hour \n", "0 -0.408106 -0.111065 -0.780079 \n", "1 -0.471788 -0.398782 1.281921 \n", "2 -0.474121 -0.102173 -0.780079 \n", "3 0.793484 0.717148 -0.780079 \n", "4 0.421837 -0.216407 -0.780079 " ] }, "execution_count": 116, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Scale the X_test data\n", "### YOUR CODE HERE ###\n", "X_test_scaled.head()" ] }, { "cell_type": "code", "execution_count": 154, "metadata": { "id": "P5nXSpRCVXq6" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "91771.77211136799\n", "0.8741432514378051 2.1564923554689073 13.476031147043757 3.670971417355869\n" ] } ], "source": [ "# Evaluate the model performance on the testing data\n", "### YOUR CODE HERE ###\n", "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n", "\n", "y_pred = model.predict(X_test_scaled)\n", "y_pred =pd.DataFrame(y_pred, columns=['fare_amount'])\n", "\n", "y_true_arr = y_test.values.ravel()\n", "y_pred_arr = y_pred.values.ravel()\n", "\n", "resid_pred = y_true_arr - y_pred_arr\n", "\n", "rss_pred = np.sum((resid_pred) ** 2)\n", "r2_pred = r2_score(y_test, y_pred)\n", "mae_pred = mean_absolute_error(y_test, y_pred)\n", "mse_pred = mean_squared_error(y_test, y_pred)\n", "rmse_pred = np.sqrt(mse_pred)\n", "print(rss_pred)\n", "print(r2_pred, mae_pred, mse_pred, rmse_pred)" ] }, { "cell_type": "code", "execution_count": 152, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([ 2.68934101, -2.6447818 , 2.55001958, ..., 4.4728535 ,\n", " 1.61532064, 1.47541722])" ] }, "execution_count": 152, "metadata": {}, "output_type": "execute_result" } ], "source": [ "resid_pred" ] }, { "cell_type": "markdown", "metadata": { "id": "L3MCKUhPJLi5" }, "source": [ "\n", "\n", "## PACE: **Execute**\n", "\n", "Consider the questions in your PACE Strategy Document to reflect on the Execute stage." ] }, { "cell_type": "markdown", "metadata": { "id": "F_l3bkxQdJ3a" }, "source": [ "### Task 9a. Results\n", "\n", "Use the code cell below to get `actual`,`predicted`, and `residual` for the testing set, and store them as columns in a `results` dataframe." ] }, { "cell_type": "code", "execution_count": 160, "metadata": { "id": "cSl5gbXfBPBN" }, "outputs": [], "source": [ "# Create a `results` dataframe\n", "### YOUR CODE HERE ###\n", "results = pd.DataFrame()\n", "results['actual'] = y_test\n", "results['predicted'] = y_pred\n", "results['residuals'] = resid_pred" ] }, { "cell_type": "code", "execution_count": 161, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
actualpredictedresiduals
012.59.8106592.689341
16.08.644782-2.644782
212.09.4499802.550020
320.519.7219340.778066
414.014.440292-0.440292
\n", "
" ], "text/plain": [ " actual predicted residuals\n", "0 12.5 9.810659 2.689341\n", "1 6.0 8.644782 -2.644782\n", "2 12.0 9.449980 2.550020\n", "3 20.5 19.721934 0.778066\n", "4 14.0 14.440292 -0.440292" ] }, "execution_count": 161, "metadata": {}, "output_type": "execute_result" } ], "source": [ "results.head()" ] }, { "cell_type": "markdown", "metadata": { "id": "mwRmSDS3eyeH" }, "source": [ "### Task 9b. Visualize model results" ] }, { "cell_type": "markdown", "metadata": { "id": "y3vQ-mB51dfd" }, "source": [ "Create a scatterplot to visualize `actual` vs. `predicted`." ] }, { "cell_type": "code", "execution_count": 163, "metadata": { "id": "IBFU_dicBjwQ" }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Create a scatterplot to visualize `predicted` over `actual`\n", "### YOUR CODE HERE ###\n", "sns.scatterplot(data=results, x='actual', y='predicted')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "id": "IbO71S_R9IcY" }, "source": [ "Visualize the distribution of the `residuals` using a histogram." ] }, { "cell_type": "code", "execution_count": 165, "metadata": { "id": "3a0UYoEr9Nx6" }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEGCAYAAACKB4k+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAYXElEQVR4nO3df5RcdZnn8fdDd9IdQqJEApshHIPHHEdwFXcyKKJzQNwBGVd0FzQzwkDMLB4BV3dcZ8h6dnb8g5U944Izy48xI0pWWN2IP4iwawZjostZhxgQkV8R5qDSJEsy7pIuNqGl08/+Ubeulaa6U0m66lZ3v1/n9Klb37r31lPV1fXpe7/3fm9kJpIkARxVdQGSpN5hKEiSSoaCJKlkKEiSSoaCJKnUX3UBR+K4447LZcuWVV2GJE0r999//z9k5uJWj03rUFi2bBnbtm2rugxJmlYi4ucTPebuI0lSyVCQJJUMBUlSyVCQJJUMBUlSyVCQJJUMBUlSyVCQJJUMBR2WzGR4eBivxyHNLIaCDkutVmPl9XdTq9WqLkXSFDIUdNj6B4+uugRJU8xQ0BFxN5I0sxgKOiLuRpJmFkNBR8zdSNLMYShIkkqGgiSpZChIkkqGgiSpZChIkkqGgiSp1PFQiIi+iPhRRNxV3F8UEfdExBPF7bFN866JiCcjYntEnNvp2iRJB+rGlsJHgcea7l8NbMrM5cCm4j4RcQqwEjgVOA+4KSL6ulCfJKnQ0VCIiKXA7wGfb2q+AFhXTK8D3tPU/pXMHMnMp4AngdM7WZ8k6UCd3lL4LPAnwFhT2wmZuROguD2+aD8ReLppvqGi7QARcXlEbIuIbbt37+5M1ZI0S3UsFCLiXcCuzLy/3UVatL1klLXMXJuZKzJzxeLFi4+oRknSgfo7uO4zgXdHxPnAILAwIm4Dno2IJZm5MyKWALuK+YeAk5qWXwrs6GB9kqRxOralkJlrMnNpZi6j3oH83cy8GNgAXFrMdilwZzG9AVgZEQMRcTKwHNjaqfokSS/VyS2FiVwLrI+I1cAvgIsAMvORiFgPPAqMAldm5v4K6pOkWasroZCZW4AtxfQvgXMmmO8a4Jpu1KRDl5nUajUWLFhQdSmSOsQzmtU2L6gjzXyGgg6JF9SRZjZDQZJUMhQkSSVDQZJUMhQkSSVDQZJUMhQkSSVDQZJUqmKYC01jmcnw8HA57Yls0sxiKOiQjI7s40PrtjI2OsL+0f1cdeu9DCxYxPDwMAsWLCCi1QjokqYLdx/pkM0ZnM+c4szmvrnzGB3Zx+q1W9xqkGYAQ0FTon/A4S+kmcBQkCSVDAVJUslQkCSVDAVJUslQkCSVDAW1pfmkNUkzl6GgttRqNVbduJH9o/urLkVSBxkKapuX4pRmPkNBklQyFCRJJUNBklQyFCRJJUNBklQyFCRJJUNBklQyFDQlGmc8Z2bVpUg6AoaCpoRXX5NmBkNBU8arr0nTn6EgSSoZCpKkkqEgSSoZCpKkkqEgSSoZCpKkkqEgSSoZCpKkUsdCISIGI2JrRPw4Ih6JiE8V7Ysi4p6IeKK4PbZpmTUR8WREbI+IcztVmySptU5uKYwAb8/MNwCnAedFxJuBq4FNmbkc2FTcJyJOAVYCpwLnATdFRF8H69MUc/wjafrrWChk3fPF3TnFTwIXAOuK9nXAe4rpC4CvZOZIZj4FPAmc3qn6NPUc/0ia/jrapxARfRHxILALuCcz7wNOyMydAMXt8cXsJwJPNy0+VLSNX+flEbEtIrbt3r27k+XrMDj+kTS9dTQUMnN/Zp4GLAVOj4jXTTJ7tFpFi3WuzcwVmbli8eLFU1WqJIkuHX2Umc8BW6j3FTwbEUsAittdxWxDwElNiy0FdnSjPk2u0Vcgaebr5NFHiyPi5cX0POAdwOPABuDSYrZLgTuL6Q3AyogYiIiTgeXA1k7Vp/bVajVW3biR/aP7qy5FUof1d3DdS4B1xRFERwHrM/OuiPgBsD4iVgO/AC4CyMxHImI98CgwClyZmX4L9Yj+waMNBWkW6FgoZOZDwBtbtP8SOGeCZa4BrulUTeq8xq6mBQsWENGqm0hSL/OMZk0pD0uVpjdDQVPOw1Kl6ctQkCSVDAVJUslQkCSVDAVJUslQkCSVDAVJUslQkCSVDAVJUslQkCSVDAVJUslQkCSVDAVJUslQkCSVDAVJUqmTV17TLNV8TWcvtiNNL24paMqNjuzjQ+u2cvHNm73YjjTNtBUKEXFmO21Sw5zB+cwZnF91GZIOUbtbCv+5zTZJ0jQ2aZ9CRJwBvAVYHBF/3PTQQqCvk4VJkrrvYB3Nc4FjivkWNLUPAxd2qihJUjUmDYXM/B7wvYi4NTN/3qWaJEkVafeQ1IGIWAssa14mM9/eiaIkSdVoNxS+Cvw18Hlgf+fKkSRVqd1QGM3MmztaiSSpcu0ekvqtiLgiIpZExKLGT0crU09oPjtZ0szX7pbCpcXtJ5raEnjV1JajXlOr1Vh140aif6DqUiR1QVuhkJknd7oQ9a7+waPZP3roXUmNrQzHP5Kmj7ZCISL+sFV7Zv6XqS1HM8noyF5Wr93CV//Nu1m4cGHV5UhqQ7u7j367aXoQOAd4ADAUNKn+gaOrLkHSIWh399FHmu9HxMuAL3WkIklSZQ536Oy9wPKpLESSVL12+xS+Rf1oI6gPhPdaYH2nipIkVaPdPoXPNE2PAj/PzKEO1CNJqlBbu4+KgfEepz5S6rHArzpZlCSpGu1eee19wFbgIuB9wH0R4dDZkjTDtLv76JPAb2fmLoCIWAx8B7ijU4VJkrqv3aOPjmoEQuGXh7CsJGmaaPeL/dsRsTEiLouIy4C7gf8+2QIRcVJEbI6IxyLikYj4aNG+KCLuiYgnittjm5ZZExFPRsT2iDj3cF+UJOnwTBoKEfHqiDgzMz8BfA54PfAG4AfA2oOsexT4eGa+FngzcGVEnAJcDWzKzOXApuI+xWMrgVOB84CbIsLrQEtSFx1sS+GzQA0gM7+emX+cmf+a+lbCZydbMDN3ZuYDxXQNeAw4EbgAWFfMtg54TzF9AfCVzBzJzKeAJ4HTD/0lSZIO18FCYVlmPjS+MTO3Ub80Z1siYhnwRuA+4ITM3FmsZydwfDHbicDTTYsNFW3j13V5RGyLiG27d+9utwRJUhsOFgqDkzw2r50niIhjgK8BH8vMya7W0mps5XxJQ+bazFyRmSsWL17cTgmSpDYdLBR+GBH/cnxjRKwG7j/YyiNiDvVAuD0zv140PxsRS4rHlwCNo5qGgJOaFl8K7DjYc0iSps7BzlP4GPCNiPgAvw6BFcBc4L2TLRj1q6rcAjyWmdc1PbSB+pXcri1u72xq/68RcR3wG9QH3Nva/kuRJB2pSUMhM58F3hIRZwOvK5rvzszvtrHuM4FLgJ9ExINF27+lHgbri62NX1A/S5rMfCQi1gOPUj9y6crMPPTLfUmSDlu711PYDGw+lBVn5r207ieA+kV6Wi1zDXDNoTyPJGnqeFayJKlkKEiSSoaCJpSZDA9PdhSxpJnGUNCEarUaq27cyP5R+/ul2cJQ0KT6B4+uugRJXWQoSJJKhoIkqWQoqKMandWZLxnGSlIPMhTUUaMj+1i9dgu1Wq3qUiS1wVBQx/UP2FktTReGgiSpZChIkkqGgiSpZCio4zwCSZo+DAV1nEcgSdOHoaCu8AgkaXowFCRJJUNBklQyFCRJJUNBklQyFCRJJUNBklQyFCRJJUNBXeOZzVLvMxTUNbVajZXX3+2ZzVIPMxTUVf2Dntks9TJDQZJUMhTUUmP/v6TZxVBQS7VajVU3bmT/6P6qS5HURYaCJuT+f2n2MRQkSSVDQV2RmR6KKk0DhoK6YnRkH1fdeq99FFKPMxTUNX1z51VdgqSDMBQkSSVDQZJUMhQkSSVDQS/RybOZHSlV6m2Ggl6ik2czj47sY/XaLR6eKvWojoVCRHwhInZFxMNNbYsi4p6IeKK4PbbpsTUR8WREbI+IcztVl9rTybOZ+wc8U1rqVZ3cUrgVOG9c29XApsxcDmwq7hMRpwArgVOLZW6KiL4O1iZJaqFjoZCZ3wf+z7jmC4B1xfQ64D1N7V/JzJHMfAp4Eji9U7VJklrrdp/CCZm5E6C4Pb5oPxF4umm+oaJNktRFvdLRHC3aWh6eEhGXR8S2iNi2e/fuDpclSbNLt0Ph2YhYAlDc7irah4CTmuZbCuxotYLMXJuZKzJzxeLFiztarCTNNt0OhQ3ApcX0pcCdTe0rI2IgIk4GlgNbu1ybJM16/Z1acUR8GTgLOC4ihoB/D1wLrI+I1cAvgIsAMvORiFgPPAqMAldmpsNpSlKXdSwUMvP3J3jonAnmvwa4plP1SJIOrlc6mjWLONSF1LsMBXWdQ11IvctQUCUc6kLqTR3rU9D007iOcqdGSB3/XI3nWbBgARGtTlWR1G2Ggkq1Wo2Lb97Miy/s7fi1lEdH9vGhdVvp6+/jtg+fzcKFCzv6fJLaYyjoAHMG5wOw//nO7++fMzifvn7HPZR6iX0KkqSSoSBJKhkKkqSSoSBJKhkKkqSSoSBJKhkKAg48mUzS7GUoCKifuLbqxo0dP2lNUm8zFFTqH3Q8Imm2MxRUKYfRlnqLoaBKjY7s5YOf28wzzzxjMEg9wFBQDwivryD1CENBPcHrK0i9wVCQJJUMBUlSyVCQJJUMBfWExqGpY2NjHqIqVchQUE8YHdnHBz+3me3bt7Py+rs9EkmqiKGgHhr3KLjq1nuJ/oGqC5FmLUNBPTXuUd/ceVWXIM1qhoIAxz2SVGcoSJJK/VUXoGr1Tn/CgZrrWrBgARFRcUXS7GAozHKN/oRe6tzNTHbs2MGfbvgpALd9+GwWLlxYcVXS7ODuI/Vcf8LoyD6uuvVejuofZM7g/KrLkWYVQ0E9qXEUktdbkLrLUFBPGx3Z67DaUhcZCrNYr3Yyj9c/cLRbDFKXGAqzVGbyzDPP9MxJa5NpdDw7/IXUeYbCDND4L/pQ/pPuxaOOJtLoeJ6oVrcipKljKMwAtVqNi2/ezMU3b57wP+nGF+eePXt47rnn2LNnT88ddTSZ5uEvxodArVZzK0KaIp6nMEM0H7rZ3FfQOL7/mWee4Yrb7+fFF/byq73PA9NvnKHMZM+ePezZs4cPfeFevvyx84kIhoeHp1XASb3MUJhBmv+DvvDaOzhqzgDrP/7Pyl1F849bypxBer4PYSKjI/u47IZvAzCwYFF5gtuLL+zlqP6B8vV7BrR0+Nx9NE212o/efPhm/+DR9M2dx44dO6ZN30E7+ubOo2/uvHEnuB19QGf0RP0r9j1IB9dzoRAR50XE9oh4MiKurrqeKrT68ppoP3rjC7Chb+68ct/6wTpop7vm3V+N10rf3Jbh0DjaqtHe2A01NjZWTjfPO/4qcBN15hs0mml6KhQiog+4EXgncArw+xFxSieea7I/5kM5mqd5PROt81DaG19e77/uLoaGhspO4VZtfQO/3hJo7BJqfDk27k+3foMj0bwF0RwOtVqt3IXWaL/w2ju46DMb2L59Oxdeewd/cMMmPnDTdxkaGmJoaIj3X3cX27dvL9/zoaEhLr55czlPI1Cafy979ux5yedg/PT4AGpX87KNsGrcb2c9rWqabD2N52t81lqFaKvP/GTPM36e5nU15hnfPtHf56G89pmm0/+IRC+9qRFxBvDnmXlucX8NQGZ+utX8K1asyG3bth3Wcw0PD/Mvrv0qX7zi3JcMtjY8PMzlt/xPANauftukg7ENDw+z6qaNfPGKcwHK6eZlmuc5WPvw8DCXXP9NjuofYHRkHwCDCxcx9uIIY2NjZRtA/0D9S7B/YB6DC1/B2OhI2YncynSb50jXMTqyj8GFi7jl8rMAWHXjxvI97B+YV87T/P7t/b+7gQPf28Z7fszipeU8/QPzuOGyt3LF5zeVv6v+gXms+8g768/V4jMBcMn132Tu/JfT19d30M9Ws+bPxQ2XvZWPr3+QF1/Yy9jYGOs+8s6DrqfV5/SvLnnLhOtpPN/Y/rEDPn/Nr3n8a124cOGEz/OvvvS/XjLPX13ylnJdjdcE8J/ed9oBz9Hq77PxXrTz2meaxvv3tasvOuzXHhH3Z+aKlo/1WChcCJyXmX9U3L8EeFNmXtU0z+XA5cXd1wDbO1zWccA/dPg5DlUv1gS9WZc1ta8X67Km9h1KXa/MzMWtHui1o49aHTJyQGpl5lpgbXfKgYjYNlGiVqUXa4LerMua2teLdVlT+6aqrp7qUwCGgJOa7i8FdlRUiyTNOr0WCj8ElkfEyRExF1gJbKi4JkmaNXpq91FmjkbEVcBGoA/4QmY+UnFZXdtVdQh6sSbozbqsqX29WJc1tW9K6uqpjmZJUrV6bfeRJKlChoIkqWQoTCAi/iIiHo+IhyLiGxHx8qbH1hTDcGyPiHO7WNNFEfFIRIxFxIpxj1VSU/HcPTE0SUR8ISJ2RcTDTW2LIuKeiHiiuD22yzWdFBGbI+Kx4nf30arriojBiNgaET8uavpU1TU11dYXET+KiLt6qKafRcRPIuLBiNjWC3VFxMsj4o7iO+qxiDhjqmoyFCZ2D/C6zHw98FNgDUAx7MZK4FTgPOCmYniObngY+OfA95sbq6ypm0OTtOFW6q+/2dXApsxcDmwq7nfTKPDxzHwt8GbgyuL9qbKuEeDtmfkG4DTgvIh4c8U1NXwUeKzpfi/UBHB2Zp7WdB5A1XX9JfDtzPxN4A3U37OpqakxRok/E/8A7wVuL6bXAGuaHtsInNHlerYAK5ruV1YTcAawcaJaKvhdLQMebrq/HVhSTC8Btlf8WboT+Ke9UhdwNPAA8Kaqa6J+XtIm4O3AXb3y+wN+Bhw3rq2yuoCFwFMUBwpNdU1uKbTng8D/KKZPBJ5uemyoaKtSlTX14vvR7ITM3AlQ3B5fVSERsQx4I3Bf1XUVu2keBHYB92Rm5TUBnwX+BBhraqu6JqiPqvC3EXF/McxO1XW9CtgNfLHY1fb5iJg/VTX11HkK3RYR3wH+UYuHPpmZdxbzfJL6LoDbG4u1mH/Kjuttp6ZWi3WypoOo8rmnjYg4Bvga8LHMHK76IkCZuR84regr+0ZEvK7KeiLiXcCuzLw/Is6qspYWzszMHRFxPHBPRDxecT39wD8BPpKZ90XEXzKFu69mdShk5jsmezwiLgXeBZyTxTYZHR6K42A1TaDK4UF6fWiSZyNiSWbujIgl1P8z7qqImEM9EG7PzK/3Sl0AmflcRGyh3hdTZU1nAu+OiPOBQWBhRNxWcU0AZOaO4nZXRHwDOL3iuoaAoWLrDuAO6qEwJTW5+2gCEXEe8KfAuzNzb9NDG4CVETEQEScDy4GtVdTYIzX1+tAkG4BLi+lLqe/T75qobxLcAjyWmdf1Ql0RsbjYQiAi5gHvAB6vsqbMXJOZSzNzGfXP0Hcz8+IqawKIiPkRsaAxDfwu9QM+qnyv/jfwdES8pmg6B3h0ymrqdqfNdPkBnqS+r/zB4uevmx77JPD31Dt23tnFmt5L/b+EEeBZDuzgraSm4rnPp36E1t9T381V1e/sy8BO4MXifVoNvIJ65+UTxe2iLtf0Vuq70x5q+iydX2VdwOuBHxU1PQz8WdFe6XvVVN9Z/Lqjuerf36uAHxc/jzQ+3z1Q12nAtuJ3+E3g2KmqyWEuJEkldx9JkkqGgiSpZChIkkqGgiSpZChIkkqGgnQEiiEGXjIAYERcFhE3HMF6nz+yyqTDM6vPaJbGK042i8wcO+jMQGb+UYdLkrrKLQXNehGxrBiT/ibqI4b+u4j4YdSvpdG41sD8iLi7uAbBwxHx/qJ9SxTXtoiIVRHx04j4HvVhGxrrvzUiLmy6/3xxe0xEbIqIB4rx+i9oUduSiPh+MZb/wxHxto6+GZr13FKQ6l4DrKJ+duiF1Me3CWBDRPwOsBjYkZm/BxARL2teuBhr5lPAbwF7gM3UzxqezAvAe7M+QN5xwN9FxIY88IzSP6B+5vo1xbUrjj7C1ylNyi0Fqe7nmfl31Me2+V3qX+gPAL9JfSypnwDviIj/GBFvy8w945Z/E7AlM3dn5q+A/9bGcwbwHyLiIeA71IccP2HcPD8EVkXEnwP/ODNrh/fypPYYClLd/ytuA/h01q+ydVpmvjozb8nMn1LfCvgJ8OmI+LMW65hozJhRir+1os9ibtH+AepbIL+VmadRH89q8IAVZn4f+B3gGeBLEfGHh/0KpTYYCtKBNgIfLK5/QEScGBHHR8RvAHsz8zbgM9THs292H3BWRLyiGCr7oqbHfkY9UAAuAOYU0y+jfg2BFyPibOCV44uJiFcW8/wN9dFWxz+vNKXsU5CaZObfRsRrgR8UF8J5HrgYeDXwFxExRn0U1g+PW25nsYvnB9RHan0AaFwn+2+AOyNiK/XRKxtbJbcD34r6xeAfpD589XhnAZ+IiBeLWtxSUEc5SqokqeTuI0lSyVCQJJUMBUlSyVCQJJUMBUlSyVCQJJUMBUlS6f8D34pBc6GbRyYAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Visualize the distribution of the `residuals`\n", "### YOUR CODE HERE ###\n", "\n", "sns.histplot(results['residuals'])\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 166, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.022208185869656265" ] }, "execution_count": 166, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Calculate residual mean\n", "### YOUR CODE HERE ###\n", "results['residuals'].mean()" ] }, { "cell_type": "markdown", "metadata": { "id": "OCnELck-9h5M" }, "source": [ "Create a scatterplot of `residuals` over `predicted`." ] }, { "cell_type": "code", "execution_count": 167, "metadata": { "id": "7Kmr2U8A95fY" }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Create a scatterplot of `residuals` over `predicted`\n", "### YOUR CODE HERE ###\n", "sns.scatterplot(data=results, x='residuals', y='predicted')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Task 9c. Coefficients\n", "\n", "Use the `coef_` attribute to get the model's coefficients. The coefficients are output in the order of the features that were used to train the model. Which feature had the greatest effect on trip fare?" ] }, { "cell_type": "code", "execution_count": 169, "metadata": {}, "outputs": [], "source": [ "# Output the model's coefficients\n", "coefficients = model.params\n" ] }, { "cell_type": "code", "execution_count": 170, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Intercept 12.674925\n", "C(rush_hour)[T.1.281920660157489] 0.183481\n", "C(RatecodeID)[T.1.1524551660464013] 7.062086\n", "C(RatecodeID)[T.2.3610460269342153] 15.154610\n", "C(RatecodeID)[T.3.569636887822029] 9.648232\n", "C(RatecodeID)[T.4.778227748709843] 23.124788\n", "C(RatecodeID)[T.118.38576867216435] 48.870687\n", "C(VendorID)[T.0.8957071444206769] -0.073360\n", "mean_distance 5.918854\n", "mean_duration 3.379850\n", "dtype: float64" ] }, "execution_count": 170, "metadata": {}, "output_type": "execute_result" } ], "source": [ "coefficients" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "What do these coefficients mean? How should they be interpreted?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The greater the coef; the greater the impact the variable have in the predicted outcome. A larger number either positive or negative has a more significant impact on the prediction." ] }, { "cell_type": "markdown", "metadata": { "id": "P6AlDDyhdzmG" }, "source": [ "### Task 9d. Conclusion\n", "\n", "1. What are the key takeaways from this notebook?\n", "\n", "\n", "\n", "2. What results can be presented from this notebook?\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As a take away, even when concidering variables with a high correlation into the model it is possible to produce strong prediction through MLR without much overfitting. MLR is very usfull not only to more accuretly predict a dependent varible, but to understand the impact of different variables on the dependent varible. The error metric can be somewhat high, but that gives as our confidence band for our results.\n", "\n", "From this notebook we can present our model with a R2 of 0.87 which explains 87% of the value of why given our selected independent variables. We can present the mean distance and duration value from trips from specific pick up and drop off location and how much those affect the fare amount. We can present a predicted fare amount based on the information available within a narrow margin of error (mae of 2.16)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Congratulations!** You've completed this lab. However, you may not notice a green check mark next to this item on Coursera's platform. Please continue your progress regardless of the check mark. Just click on the \"save\" icon at the top of this notebook to ensure your work has been logged. " ] } ], "metadata": { "colab": { "provenance": [], "toc_visible": true }, "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": "3.7.6" } }, "nbformat": 4, "nbformat_minor": 1 }