{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Freight Optimization\n", "\n", "Notes: \n", "\n", "[1 - Introduction](01-intro.html) | **Freight Optimization** " ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Datasets " ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "url = \"https://notes.pipal.in/2018/vmware-ml2/10-Secondary-Freight-Data.csv\"\n", "matrix_url = \"https://notes.pipal.in/2018/vmware-ml2/13-location-distance-matrix.csv\"" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "freight = pd.read_csv(url)" ] }, { "cell_type": "code", "execution_count": 5, "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", "
Sales Order No. NewDateSupplying DC LocationCustomer Code NewCustomer TownQty (in cases)Freight/CartageRemarksAmount
020049120141-Jul-08Pune190886PUNE15.05Charges - Per case75
120049128461-Jul-08Ahemadabad190406JAMNAGAR30.018Charges - Per case540
220049134181-Jul-08Jaipur188582BAGRU6.020Charges - Per case120
320049164502-Jul-08Raipur191024RAIPUR23.05Charges - Per case104
420049168062-Jul-08Zirakhpur207786Banur1.01,000Charges - Per Consignment1,000
\n", "
" ], "text/plain": [ " Sales Order No. New Date Supplying DC Location Customer Code New \\\n", "0 2004912014 1-Jul-08 Pune 190886 \n", "1 2004912846 1-Jul-08 Ahemadabad 190406 \n", "2 2004913418 1-Jul-08 Jaipur 188582 \n", "3 2004916450 2-Jul-08 Raipur 191024 \n", "4 2004916806 2-Jul-08 Zirakhpur 207786 \n", "\n", " Customer Town Qty (in cases) Freight/Cartage Remarks \\\n", "0 PUNE 15.0 5 Charges - Per case \n", "1 JAMNAGAR 30.0 18 Charges - Per case \n", "2 BAGRU 6.0 20 Charges - Per case \n", "3 RAIPUR 23.0 5 Charges - Per case \n", "4 Banur 1.0 1,000 Charges - Per Consignment \n", "\n", " Amount \n", "0 75 \n", "1 540 \n", "2 120 \n", "3 104 \n", "4 1,000 " ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.head()" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Sales Order No. New int64\n", "Date object\n", "Supplying DC Location object\n", "Customer Code New int64\n", "Customer Town object\n", "Qty (in cases) float64\n", "Freight/Cartage object\n", "Remarks object\n", "Amount object\n", "dtype: object" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.dtypes" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Index(['Sales Order No. New', 'Date', 'Supplying DC Location',\n", " 'Customer Code New', 'Customer Town', 'Qty (in cases)',\n", " 'Freight/Cartage', 'Remarks', 'Amount'],\n", " dtype='object')" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.columns" ] }, { "cell_type": "code", "execution_count": 90, "metadata": {}, "outputs": [], "source": [ "matrix = pd.read_csv(matrix_url)" ] }, { "cell_type": "code", "execution_count": 91, "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", "
SourceS. LongitudeS. LatitudeDestinationD. LongitudeD. LatitudeLaneDistance
0BANGALORE77.5912.98BANGALORE77.5912.98BANGALORE to BANGALORE0.00
1BANGALORE77.5912.98CHENNAI80.2413.07BANGALORE to CHENNAI294.47
2BANGALORE77.5912.98MUMBAI72.8418.98BANGALORE to MUMBAI554.79
3BANGALORE77.5912.98HYDERABAD78.4917.39BANGALORE to HYDERABAD142.09
4BANGALORE77.5912.98NEW DELHI77.1728.62BANGALORE to NEW DELHI381.46
\n", "
" ], "text/plain": [ " Source S. Longitude S. Latitude Destination D. Longitude \\\n", "0 BANGALORE 77.59 12.98 BANGALORE 77.59 \n", "1 BANGALORE 77.59 12.98 CHENNAI 80.24 \n", "2 BANGALORE 77.59 12.98 MUMBAI 72.84 \n", "3 BANGALORE 77.59 12.98 HYDERABAD 78.49 \n", "4 BANGALORE 77.59 12.98 NEW DELHI 77.17 \n", "\n", " D. Latitude Lane Distance \n", "0 12.98 BANGALORE to BANGALORE 0.00 \n", "1 13.07 BANGALORE to CHENNAI 294.47 \n", "2 18.98 BANGALORE to MUMBAI 554.79 \n", "3 17.39 BANGALORE to HYDERABAD 142.09 \n", "4 28.62 BANGALORE to NEW DELHI 381.46 " ] }, "execution_count": 91, "metadata": {}, "output_type": "execute_result" } ], "source": [ "matrix.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Refine the Data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Basic Cleaning:**\n", "* fix column names\n", "* drop unused/redundant columns" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Basic Cleaning" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "columns = {\n", " 'Sales Order No. New': 'orderno',\n", " 'Date': 'date', \n", " 'Supplying DC Location': 'source',\n", " 'Customer Code New': 'custcode',\n", " 'Customer Town': 'dest', \n", " 'Qty (in cases)': 'qty',\n", " 'Freight/Cartage': 'cartage', \n", " 'Remarks': 'remarks', \n", " 'Amount': 'amount'\n", "}" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "freight.rename(columns=columns, inplace=True)" ] }, { "cell_type": "code", "execution_count": 13, "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", "
ordernodatesourcecustcodedestqtycartageremarksamount
020049120141-Jul-08Pune190886PUNE15.05Charges - Per case75
120049128461-Jul-08Ahemadabad190406JAMNAGAR30.018Charges - Per case540
220049134181-Jul-08Jaipur188582BAGRU6.020Charges - Per case120
320049164502-Jul-08Raipur191024RAIPUR23.05Charges - Per case104
420049168062-Jul-08Zirakhpur207786Banur1.01,000Charges - Per Consignment1,000
\n", "
" ], "text/plain": [ " orderno date source custcode dest qty cartage \\\n", "0 2004912014 1-Jul-08 Pune 190886 PUNE 15.0 5 \n", "1 2004912846 1-Jul-08 Ahemadabad 190406 JAMNAGAR 30.0 18 \n", "2 2004913418 1-Jul-08 Jaipur 188582 BAGRU 6.0 20 \n", "3 2004916450 2-Jul-08 Raipur 191024 RAIPUR 23.0 5 \n", "4 2004916806 2-Jul-08 Zirakhpur 207786 Banur 1.0 1,000 \n", "\n", " remarks amount \n", "0 Charges - Per case 75 \n", "1 Charges - Per case 540 \n", "2 Charges - Per case 120 \n", "3 Charges - Per case 104 \n", "4 Charges - Per Consignment 1,000 " ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.head()" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "freight.drop([\"orderno\", \"custcode\"], axis=1, inplace=True)" ] }, { "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", "
datesourcedestqtycartageremarksamount
01-Jul-08PunePUNE15.05Charges - Per case75
11-Jul-08AhemadabadJAMNAGAR30.018Charges - Per case540
21-Jul-08JaipurBAGRU6.020Charges - Per case120
32-Jul-08RaipurRAIPUR23.05Charges - Per case104
42-Jul-08ZirakhpurBanur1.01,000Charges - Per Consignment1,000
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks \\\n", "0 1-Jul-08 Pune PUNE 15.0 5 Charges - Per case \n", "1 1-Jul-08 Ahemadabad JAMNAGAR 30.0 18 Charges - Per case \n", "2 1-Jul-08 Jaipur BAGRU 6.0 20 Charges - Per case \n", "3 2-Jul-08 Raipur RAIPUR 23.0 5 Charges - Per case \n", "4 2-Jul-08 Zirakhpur Banur 1.0 1,000 Charges - Per Consignment \n", "\n", " amount \n", "0 75 \n", "1 540 \n", "2 120 \n", "3 104 \n", "4 1,000 " ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.head()" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "dtype('O')" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.amount.dtype" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array(['75', '540', '120', '104', '1,000', '84', '255', '138', '48', '39',\n", " '380', '54', '42', '65', '57', '38', '113', '125', '119', '26'], dtype=object)" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.amount.unique()[:20]" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "date object\n", "source object\n", "dest object\n", "qty float64\n", "cartage object\n", "remarks object\n", "amount object\n", "dtype: object" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.dtypes" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 75\n", "1 540\n", "2 120\n", "3 104\n", "4 1,000\n", "Name: amount, dtype: object" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# remove commas in the numbers\n", "freight.amount.head()" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 75\n", "1 540\n", "2 120\n", "3 104\n", "4 1000\n", "Name: amount, dtype: object" ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.amount.head().str.replace(\",\", \"\")" ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [], "source": [ "freight.amount = freight.amount.str.replace(\",\", \"\")" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 75\n", "1 540\n", "2 120\n", "3 104\n", "4 1000\n", "Name: amount, dtype: object" ] }, "execution_count": 36, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.amount.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Problem:** Replace commas from `cartage` column." ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [], "source": [ "freight.cartage = freight.cartage.str.replace(\",\", \"\")" ] }, { "cell_type": "code", "execution_count": 42, "metadata": {}, "outputs": [], "source": [ "# fix column types\n", "freight.amount = pd.to_numeric(freight.amount)\n", "freight.cartage = pd.to_numeric(freight.cartage)\n", "freight.date = pd.to_datetime(freight.date)" ] }, { "cell_type": "code", "execution_count": 43, "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", "
datesourcedestqtycartageremarksamount
02008-07-01PunePUNE15.05Charges - Per case75
12008-07-01AhemadabadJAMNAGAR30.018Charges - Per case540
22008-07-01JaipurBAGRU6.020Charges - Per case120
32008-07-02RaipurRAIPUR23.05Charges - Per case104
42008-07-02ZirakhpurBanur1.01000Charges - Per Consignment1000
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks \\\n", "0 2008-07-01 Pune PUNE 15.0 5 Charges - Per case \n", "1 2008-07-01 Ahemadabad JAMNAGAR 30.0 18 Charges - Per case \n", "2 2008-07-01 Jaipur BAGRU 6.0 20 Charges - Per case \n", "3 2008-07-02 Raipur RAIPUR 23.0 5 Charges - Per case \n", "4 2008-07-02 Zirakhpur Banur 1.0 1000 Charges - Per Consignment \n", "\n", " amount \n", "0 75 \n", "1 540 \n", "2 120 \n", "3 104 \n", "4 1000 " ] }, "execution_count": 43, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.head()" ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "date datetime64[ns]\n", "source object\n", "dest object\n", "qty float64\n", "cartage int64\n", "remarks object\n", "amount int64\n", "dtype: object" ] }, "execution_count": 44, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# dtypes\n", "freight.dtypes" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Q:** Is the `remarks` column required?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Lets see." ] }, { "cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array(['Charges - Per case', 'Charges - Per Consignment'], dtype=object)" ] }, "execution_count": 45, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.remarks.unique()" ] }, { "cell_type": "code", "execution_count": 46, "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", "
datesourcedestqtycartageremarksamount
02008-07-01PunePUNE15.05Charges - Per case75
12008-07-01AhemadabadJAMNAGAR30.018Charges - Per case540
22008-07-01JaipurBAGRU6.020Charges - Per case120
32008-07-02RaipurRAIPUR23.05Charges - Per case104
52008-07-02RaipurSARIPALI (C.G.)4.021Charges - Per case84
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks \\\n", "0 2008-07-01 Pune PUNE 15.0 5 Charges - Per case \n", "1 2008-07-01 Ahemadabad JAMNAGAR 30.0 18 Charges - Per case \n", "2 2008-07-01 Jaipur BAGRU 6.0 20 Charges - Per case \n", "3 2008-07-02 Raipur RAIPUR 23.0 5 Charges - Per case \n", "5 2008-07-02 Raipur SARIPALI (C.G.) 4.0 21 Charges - Per case \n", "\n", " amount \n", "0 75 \n", "1 540 \n", "2 120 \n", "3 104 \n", "5 84 " ] }, "execution_count": 46, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight[freight.remarks=='Charges - Per case'].head()" ] }, { "cell_type": "code", "execution_count": 47, "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", "
datesourcedestqtycartageremarksamount
42008-07-02ZirakhpurBanur1.01000Charges - Per Consignment1000
252008-07-03DelhiNEW DELHI1.01000Charges - Per Consignment1000
312008-07-03GhaziabadNOIDA17.0250Charges - Per Consignment250
322008-07-03ChennaiCHENNAI4.0550Charges - Per Consignment550
382008-07-04KolkataKOLKATA11.0750Charges - Per Consignment750
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks \\\n", "4 2008-07-02 Zirakhpur Banur 1.0 1000 Charges - Per Consignment \n", "25 2008-07-03 Delhi NEW DELHI 1.0 1000 Charges - Per Consignment \n", "31 2008-07-03 Ghaziabad NOIDA 17.0 250 Charges - Per Consignment \n", "32 2008-07-03 Chennai CHENNAI 4.0 550 Charges - Per Consignment \n", "38 2008-07-04 Kolkata KOLKATA 11.0 750 Charges - Per Consignment \n", "\n", " amount \n", "4 1000 \n", "25 1000 \n", "31 250 \n", "32 550 \n", "38 750 " ] }, "execution_count": 47, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight[freight.remarks=='Charges - Per Consignment'].head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Fix Missing Values" ] }, { "cell_type": "code", "execution_count": 49, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "60976" ] }, "execution_count": 49, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(freight)" ] }, { "cell_type": "code", "execution_count": 50, "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", "
datesourcedestqtycartageremarksamount
0FalseFalseFalseFalseFalseFalseFalse
1FalseFalseFalseFalseFalseFalseFalse
2FalseFalseFalseFalseFalseFalseFalse
3FalseFalseFalseFalseFalseFalseFalse
4FalseFalseFalseFalseFalseFalseFalse
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount\n", "0 False False False False False False False\n", "1 False False False False False False False\n", "2 False False False False False False False\n", "3 False False False False False False False\n", "4 False False False False False False False" ] }, "execution_count": 50, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.isnull().head()" ] }, { "cell_type": "code", "execution_count": 51, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "date 0\n", "source 6\n", "dest 0\n", "qty 0\n", "cartage 0\n", "remarks 0\n", "amount 0\n", "dtype: int64" ] }, "execution_count": 51, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.isnull().sum()" ] }, { "cell_type": "code", "execution_count": 52, "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", "
datesourcedestqtycartageremarksamount
285722008-12-29NaNLEH120.00Charges - Per case0
343772009-01-31NaNLEH109.00Charges - Per case0
452042009-03-31NaNLEH15.00Charges - Per case0
528522009-05-18NaNLEH135.00Charges - Per case0
534382009-05-22NaNLEH160.00Charges - Per case0
609032009-06-30NaNLEH154.00Charges - Per case0
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount\n", "28572 2008-12-29 NaN LEH 120.0 0 Charges - Per case 0\n", "34377 2009-01-31 NaN LEH 109.0 0 Charges - Per case 0\n", "45204 2009-03-31 NaN LEH 15.0 0 Charges - Per case 0\n", "52852 2009-05-18 NaN LEH 135.0 0 Charges - Per case 0\n", "53438 2009-05-22 NaN LEH 160.0 0 Charges - Per case 0\n", "60903 2009-06-30 NaN LEH 154.0 0 Charges - Per case 0" ] }, "execution_count": 52, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight[freight.source.isnull()]" ] }, { "cell_type": "code", "execution_count": 53, "metadata": {}, "outputs": [], "source": [ "# remove NaN values\n", "freight.dropna(inplace=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Standadize text fields" ] }, { "cell_type": "code", "execution_count": 58, "metadata": {}, "outputs": [], "source": [ "shortunits = {\n", " \"Charges - Per case\": \"case\",\n", " \"Charges - Per Consignment\": \"consignment\"\n", "}\n", "freight.remarks = freight.remarks.replace(shortunits)" ] }, { "cell_type": "code", "execution_count": 59, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "case 57776\n", "consignment 3194\n", "Name: remarks, dtype: int64" ] }, "execution_count": 59, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.remarks.value_counts()" ] }, { "cell_type": "code", "execution_count": 61, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Chennai 1138\n", "BANGALORE 1134\n", "MUMBAI 1014\n", "AHMEDABAD 799\n", "CHENNAI 779\n", "KOLKATA 748\n", "NEW DELHI 721\n", "HYDERABAD 620\n", "PUNE 601\n", "Hyderabad 563\n", "THANE 506\n", "New Delhi 469\n", "RANCHI 462\n", "Mumbai 424\n", "LUCKNOW 414\n", "SALEM 382\n", "MADURAI 345\n", "PATNA 342\n", "SURAT 292\n", "JAIPUR 264\n", "RAJKOT 252\n", "Bangalore 245\n", "VARANASI 241\n", "TIRUNELVELI 236\n", "JAMSHEDPUR 226\n", "KANPUR 211\n", "TRIVANDRUM 209\n", "JABALPUR 178\n", "THRISSUR 174\n", "AGRA 168\n", " ... \n", "BHUPALPALLY 1\n", "Bachepalli 1\n", "Dalsingsarai 1\n", "Bellampalle 1\n", "Thuvarankurichy 1\n", "Distt:Ahmedabad 1\n", "Ghumarwin 1\n", "Puttur 1\n", "Banga 1\n", "LALSOT 1\n", "BANTHARA BAZAAR, 1\n", "KARANPRAYAG 1\n", "DAMANDIU 1\n", "DEBAI 1\n", "TAMKUHI ROAD 1\n", "TIRODA 1\n", "Ramanagaram 1\n", "DUNGANJ 1\n", "BARSHI 1\n", "ANPARA BAZAR 1\n", "Namakkal 1\n", "SONARPUR 1\n", "JAMUI 1\n", "Bahadurgarh 1\n", "ANDOLE 1\n", "SHAJAPUR 1\n", " RAMGANJ MANDI 1\n", "NAWABGANJ 1\n", "Lanka 1\n", "CHINTPURNI 1\n", "Name: dest, Length: 1923, dtype: int64" ] }, "execution_count": 61, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.dest.value_counts()" ] }, { "cell_type": "code", "execution_count": 72, "metadata": {}, "outputs": [], "source": [ "def fix_names(c):\n", " c = c.str.strip()\n", " c = c.str.title()\n", " return c" ] }, { "cell_type": "code", "execution_count": 73, "metadata": {}, "outputs": [], "source": [ "freight.source = fix_names(freight.source)\n", "freight.dest = fix_names(freight.dest)" ] }, { "cell_type": "code", "execution_count": 74, "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", "
datesourcedestqtycartageremarksamount
02008-07-01PunePune15.05case75
12008-07-01AhemadabadJamnagar30.018case540
22008-07-01JaipurBagru6.020case120
32008-07-02RaipurRaipur23.05case104
42008-07-02ZirakhpurBanur1.01000consignment1000
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount\n", "0 2008-07-01 Pune Pune 15.0 5 case 75\n", "1 2008-07-01 Ahemadabad Jamnagar 30.0 18 case 540\n", "2 2008-07-01 Jaipur Bagru 6.0 20 case 120\n", "3 2008-07-02 Raipur Raipur 23.0 5 case 104\n", "4 2008-07-02 Zirakhpur Banur 1.0 1000 consignment 1000" ] }, "execution_count": 74, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Q:** Which source has the maximum quantity of transfer? " ] }, { "cell_type": "code", "execution_count": 77, "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", "
qtycartageamount
source
Ahemadabad67593.02855601692459
Bangalore221221.02942493254982
Bbsr44318.0545201099986
Bhiwandi103646.02524681670543
Chandigarh4247.054521235
\n", "
" ], "text/plain": [ " qty cartage amount\n", "source \n", "Ahemadabad 67593.0 285560 1692459\n", "Bangalore 221221.0 294249 3254982\n", "Bbsr 44318.0 54520 1099986\n", "Bhiwandi 103646.0 252468 1670543\n", "Chandigarh 4247.0 545 21235" ] }, "execution_count": 77, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.groupby('source').sum().head()" ] }, { "cell_type": "code", "execution_count": 81, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "source\n", "Bangalore 221221.0\n", "Chennai 140902.0\n", "Madurai 133278.0\n", "Hyderabad 122391.0\n", "Cochin 103790.0\n", "Name: qty, dtype: float64" ] }, "execution_count": 81, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.groupby('source').sum().qty.sort_values(ascending=False).head()" ] }, { "cell_type": "code", "execution_count": 82, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "source\n", "Bangalore 221221.0\n", "Chennai 140902.0\n", "Madurai 133278.0\n", "Hyderabad 122391.0\n", "Cochin 103790.0\n", "Name: qty, dtype: float64" ] }, "execution_count": 82, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(freight.groupby('source')\n", " .sum()\n", " .qty\n", " .sort_values(ascending=False)\n", " .head()\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Q:** What is the source/destination pair that has the maximum quantity of transfer? " ] }, { "cell_type": "code", "execution_count": 84, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "source dest \n", "Bangalore Bangalore 111120.0\n", "Chennai Chennai 80251.0\n", "Bhiwandi Mumbai 58666.0\n", "Hyderabad Hyderabad 54372.0\n", "Delhi New Delhi 51669.0\n", "Name: qty, dtype: float64" ] }, "execution_count": 84, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(freight.groupby(['source', 'dest'])\n", " .sum()\n", " .qty\n", " .sort_values(ascending=False)\n", " .head()\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Q:** Show data for Bangalore -> Bangalore." ] }, { "cell_type": "code", "execution_count": 86, "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", "
datesourcedestqtycartageremarksamount
362008-07-04BangaloreBangalore345.07case2415
512008-07-04BangaloreBangalore54.07case378
1212008-07-04BangaloreBangalore126.07case882
1252008-07-04BangaloreBangalore40.07case280
1502008-07-05BangaloreBangalore76.07case532
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount\n", "36 2008-07-04 Bangalore Bangalore 345.0 7 case 2415\n", "51 2008-07-04 Bangalore Bangalore 54.0 7 case 378\n", "121 2008-07-04 Bangalore Bangalore 126.0 7 case 882\n", "125 2008-07-04 Bangalore Bangalore 40.0 7 case 280\n", "150 2008-07-05 Bangalore Bangalore 76.0 7 case 532" ] }, "execution_count": 86, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight[(freight.source == 'Bangalore') & (freight.dest == 'Bangalore')].head()" ] }, { "cell_type": "code", "execution_count": 88, "metadata": {}, "outputs": [], "source": [ "freight['computed_amount'] = freight.qty * freight.cartage" ] }, { "cell_type": "code", "execution_count": 89, "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", "
datesourcedestqtycartageremarksamountcomputed_amount
02008-07-01PunePune15.05case7575.0
12008-07-01AhemadabadJamnagar30.018case540540.0
22008-07-01JaipurBagru6.020case120120.0
32008-07-02RaipurRaipur23.05case104115.0
42008-07-02ZirakhpurBanur1.01000consignment10001000.0
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount \\\n", "0 2008-07-01 Pune Pune 15.0 5 case 75 \n", "1 2008-07-01 Ahemadabad Jamnagar 30.0 18 case 540 \n", "2 2008-07-01 Jaipur Bagru 6.0 20 case 120 \n", "3 2008-07-02 Raipur Raipur 23.0 5 case 104 \n", "4 2008-07-02 Zirakhpur Banur 1.0 1000 consignment 1000 \n", "\n", " computed_amount \n", "0 75.0 \n", "1 540.0 \n", "2 120.0 \n", "3 115.0 \n", "4 1000.0 " ] }, "execution_count": 89, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Problem:** Refine the location-distance-matrix data and merge it with the freight dataframe.\n", "\n", "Hint: `pd.merge?`" ] }, { "cell_type": "code", "execution_count": 94, "metadata": {}, "outputs": [], "source": [ "matrix = pd.read_csv(matrix_url)" ] }, { "cell_type": "code", "execution_count": 95, "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", "
SourceS. LongitudeS. LatitudeDestinationD. LongitudeD. LatitudeLaneDistance
0BANGALORE77.5912.98BANGALORE77.5912.98BANGALORE to BANGALORE0.00
1BANGALORE77.5912.98CHENNAI80.2413.07BANGALORE to CHENNAI294.47
2BANGALORE77.5912.98MUMBAI72.8418.98BANGALORE to MUMBAI554.79
3BANGALORE77.5912.98HYDERABAD78.4917.39BANGALORE to HYDERABAD142.09
4BANGALORE77.5912.98NEW DELHI77.1728.62BANGALORE to NEW DELHI381.46
\n", "
" ], "text/plain": [ " Source S. Longitude S. Latitude Destination D. Longitude \\\n", "0 BANGALORE 77.59 12.98 BANGALORE 77.59 \n", "1 BANGALORE 77.59 12.98 CHENNAI 80.24 \n", "2 BANGALORE 77.59 12.98 MUMBAI 72.84 \n", "3 BANGALORE 77.59 12.98 HYDERABAD 78.49 \n", "4 BANGALORE 77.59 12.98 NEW DELHI 77.17 \n", "\n", " D. Latitude Lane Distance \n", "0 12.98 BANGALORE to BANGALORE 0.00 \n", "1 13.07 BANGALORE to CHENNAI 294.47 \n", "2 18.98 BANGALORE to MUMBAI 554.79 \n", "3 17.39 BANGALORE to HYDERABAD 142.09 \n", "4 28.62 BANGALORE to NEW DELHI 381.46 " ] }, "execution_count": 95, "metadata": {}, "output_type": "execute_result" } ], "source": [ "matrix.head()" ] }, { "cell_type": "code", "execution_count": 97, "metadata": {}, "outputs": [], "source": [ "# fix column names\n", "columns = {\n", " \"Source\": \"source\",\n", " \"S. Longitude\": \"slon\",\n", " \"S. Latitude\": \"slat\",\n", " \"Destination\": \"dest\",\n", " \"D. Longitude\": \"dlon\",\n", " \"D. Latitude\": \"dlat\",\n", " \"Lane\": \"lane\",\n", " \"Distance\": \"distance\"\n", "} \n", "matrix.rename(columns=columns, inplace=True)" ] }, { "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", "
sourceslonslatdestdlondlatlanedistance
0BANGALORE77.5912.98BANGALORE77.5912.98BANGALORE to BANGALORE0.00
1BANGALORE77.5912.98CHENNAI80.2413.07BANGALORE to CHENNAI294.47
2BANGALORE77.5912.98MUMBAI72.8418.98BANGALORE to MUMBAI554.79
3BANGALORE77.5912.98HYDERABAD78.4917.39BANGALORE to HYDERABAD142.09
4BANGALORE77.5912.98NEW DELHI77.1728.62BANGALORE to NEW DELHI381.46
\n", "
" ], "text/plain": [ " source slon slat dest dlon dlat lane \\\n", "0 BANGALORE 77.59 12.98 BANGALORE 77.59 12.98 BANGALORE to BANGALORE \n", "1 BANGALORE 77.59 12.98 CHENNAI 80.24 13.07 BANGALORE to CHENNAI \n", "2 BANGALORE 77.59 12.98 MUMBAI 72.84 18.98 BANGALORE to MUMBAI \n", "3 BANGALORE 77.59 12.98 HYDERABAD 78.49 17.39 BANGALORE to HYDERABAD \n", "4 BANGALORE 77.59 12.98 NEW DELHI 77.17 28.62 BANGALORE to NEW DELHI \n", "\n", " distance \n", "0 0.00 \n", "1 294.47 \n", "2 554.79 \n", "3 142.09 \n", "4 381.46 " ] }, "execution_count": 98, "metadata": {}, "output_type": "execute_result" } ], "source": [ "matrix.head()" ] }, { "cell_type": "code", "execution_count": 99, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "source 0\n", "slon 0\n", "slat 0\n", "dest 0\n", "dlon 0\n", "dlat 0\n", "lane 0\n", "distance 0\n", "dtype: int64" ] }, "execution_count": 99, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# missing values\n", "matrix.isnull().sum()" ] }, { "cell_type": "code", "execution_count": 101, "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", "
sourceslonslatdestdlondlatlanedistance
0Bangalore77.5912.98Bangalore77.5912.98BANGALORE to BANGALORE0.00
1Bangalore77.5912.98Chennai80.2413.07BANGALORE to CHENNAI294.47
2Bangalore77.5912.98Mumbai72.8418.98BANGALORE to MUMBAI554.79
3Bangalore77.5912.98Hyderabad78.4917.39BANGALORE to HYDERABAD142.09
4Bangalore77.5912.98New Delhi77.1728.62BANGALORE to NEW DELHI381.46
\n", "
" ], "text/plain": [ " source slon slat dest dlon dlat lane \\\n", "0 Bangalore 77.59 12.98 Bangalore 77.59 12.98 BANGALORE to BANGALORE \n", "1 Bangalore 77.59 12.98 Chennai 80.24 13.07 BANGALORE to CHENNAI \n", "2 Bangalore 77.59 12.98 Mumbai 72.84 18.98 BANGALORE to MUMBAI \n", "3 Bangalore 77.59 12.98 Hyderabad 78.49 17.39 BANGALORE to HYDERABAD \n", "4 Bangalore 77.59 12.98 New Delhi 77.17 28.62 BANGALORE to NEW DELHI \n", "\n", " distance \n", "0 0.00 \n", "1 294.47 \n", "2 554.79 \n", "3 142.09 \n", "4 381.46 " ] }, "execution_count": 101, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# standadize names\n", "matrix.source = fix_names(matrix.source)\n", "matrix.dest = fix_names(matrix.dest)\n", "matrix.head()" ] }, { "cell_type": "code", "execution_count": 111, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(65890, 65896)" ] }, "execution_count": 111, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(matrix.source + \"--\" + matrix.dest).nunique(), len(matrix)" ] }, { "cell_type": "code", "execution_count": 113, "metadata": {}, "outputs": [], "source": [ "# drop the duplicates\n", "matrix.drop_duplicates(['source', 'dest'], inplace=True)" ] }, { "cell_type": "code", "execution_count": 116, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "65890" ] }, "execution_count": 116, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(matrix)" ] }, { "cell_type": "code", "execution_count": 114, "metadata": {}, "outputs": [], "source": [ "df = pd.merge(freight, matrix, how='left', on=['source', 'dest'])" ] }, { "cell_type": "code", "execution_count": 115, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(60970, 65890, 60970)" ] }, "execution_count": 115, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(freight), len(matrix), len(df)" ] }, { "cell_type": "code", "execution_count": 117, "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", "
datesourcedestqtycartageremarksamountcomputed_amountslonslatdlondlatlanedistance
02008-07-01PunePune15.05case7575.073.8518.5373.8518.53PUNE to PUNE0.00
12008-07-01AhemadabadJamnagar30.018case540540.0NaNNaNNaNNaNNaNNaN
22008-07-01JaipurBagru6.020case120120.075.8126.9275.5326.82JAIPUR to BAGRU30.66
32008-07-02RaipurRaipur23.05case104115.081.6521.2381.6521.23RAIPUR to RAIPUR0.00
42008-07-02ZirakhpurBanur1.01000consignment10001000.0NaNNaNNaNNaNNaNNaN
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount \\\n", "0 2008-07-01 Pune Pune 15.0 5 case 75 \n", "1 2008-07-01 Ahemadabad Jamnagar 30.0 18 case 540 \n", "2 2008-07-01 Jaipur Bagru 6.0 20 case 120 \n", "3 2008-07-02 Raipur Raipur 23.0 5 case 104 \n", "4 2008-07-02 Zirakhpur Banur 1.0 1000 consignment 1000 \n", "\n", " computed_amount slon slat dlon dlat lane distance \n", "0 75.0 73.85 18.53 73.85 18.53 PUNE to PUNE 0.00 \n", "1 540.0 NaN NaN NaN NaN NaN NaN \n", "2 120.0 75.81 26.92 75.53 26.82 JAIPUR to BAGRU 30.66 \n", "3 115.0 81.65 21.23 81.65 21.23 RAIPUR to RAIPUR 0.00 \n", "4 1000.0 NaN NaN NaN NaN NaN NaN " ] }, "execution_count": 117, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head()" ] }, { "cell_type": "code", "execution_count": 121, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "date 0\n", "source 0\n", "dest 0\n", "qty 0\n", "cartage 0\n", "remarks 0\n", "amount 0\n", "computed_amount 0\n", "slon 15782\n", "slat 15782\n", "dlon 15782\n", "dlat 15782\n", "lane 15782\n", "distance 15782\n", "dtype: int64" ] }, "execution_count": 121, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.isnull().sum()" ] }, { "cell_type": "code", "execution_count": 122, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "60970" ] }, "execution_count": 122, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(df)" ] }, { "cell_type": "code", "execution_count": 133, "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", "
datesourcedestqtycartageremarksamountcomputed_amountslonslatdlondlatlanedistance
12008-07-01AhemadabadJamnagar30.018case540540.0NaNNaNNaNNaNNaNNaN
42008-07-02ZirakhpurBanur1.01000consignment10001000.0NaNNaNNaNNaNNaNNaN
112008-07-02CochinMattanchery38.010case380380.0NaNNaNNaNNaNNaNNaN
312008-07-03GhaziabadNoida17.0250consignment2504250.0NaNNaNNaNNaNNaNNaN
372008-07-04AhemadabadDerol7.024case168168.0NaNNaNNaNNaNNaNNaN
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount \\\n", "1 2008-07-01 Ahemadabad Jamnagar 30.0 18 case 540 \n", "4 2008-07-02 Zirakhpur Banur 1.0 1000 consignment 1000 \n", "11 2008-07-02 Cochin Mattanchery 38.0 10 case 380 \n", "31 2008-07-03 Ghaziabad Noida 17.0 250 consignment 250 \n", "37 2008-07-04 Ahemadabad Derol 7.0 24 case 168 \n", "\n", " computed_amount slon slat dlon dlat lane distance \n", "1 540.0 NaN NaN NaN NaN NaN NaN \n", "4 1000.0 NaN NaN NaN NaN NaN NaN \n", "11 380.0 NaN NaN NaN NaN NaN NaN \n", "31 4250.0 NaN NaN NaN NaN NaN NaN \n", "37 168.0 NaN NaN NaN NaN NaN NaN " ] }, "execution_count": 133, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df.isnull().sum(axis=1)>0].head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Q:** What is the total amount where distance is missing?" ] }, { "cell_type": "code", "execution_count": 134, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "11070970" ] }, "execution_count": 134, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df.isnull().sum(axis=1)>0].amount.sum()" ] }, { "cell_type": "code", "execution_count": 137, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.271028672517011" ] }, "execution_count": 137, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df.isnull().sum(axis=1)>0].amount.sum() / df.amount.sum()" ] }, { "cell_type": "code", "execution_count": 138, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(31, 54, 31)" ] }, "execution_count": 138, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.source.nunique(), matrix.source.nunique(), df.source.nunique()" ] }, { "cell_type": "code", "execution_count": 139, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array(['Pune', 'Ahemadabad', 'Jaipur', 'Raipur', 'Zirakhpur', 'Cochin',\n", " 'Rohtak', 'Chennai', 'Delhi', 'Ghaziabad', 'Kolkata', 'Bangalore',\n", " 'Madurai', 'Vijaywada', 'Nagpur', 'Lucknow', 'Coimbatore', 'Goa',\n", " 'Hyderabad', 'Ranchi', 'Bhiwandi', 'Gauwhati', 'Bbsr', 'Varanasi',\n", " 'Patna', 'Indore', 'Rishikesh', 'Jammu', 'Parwanoo', 'Chandigarh',\n", " 'Haldwani'], dtype=object)" ] }, "execution_count": 139, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.source.unique()" ] }, { "cell_type": "code", "execution_count": 140, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array(['Bangalore', 'Chennai', 'Mumbai', 'Hyderabad', 'New Delhi', 'Salem',\n", " 'Coimbatore', 'Pune', 'Srinagar', 'Madurai', 'Kolkata', 'Raipur',\n", " 'Lucknow', 'Imphal', 'Tirunelveli', 'Secunderabad', 'Ahmedabad',\n", " 'Jaipur', 'Mangalore', 'Dimapur', 'Thane', 'Ranchi', 'Kanpur',\n", " 'Varanasi', 'Kollam', 'Vishakapatnam', 'Mysore', 'Trichy', 'Hubli',\n", " 'Madanapalle', 'Bareilly', 'Pudukottai', 'Barhampur', 'Tirupur',\n", " 'Ludhiana', 'Jamshedpur', 'Bhubaneshwar', 'Bhiwandi', 'Chandigarh',\n", " 'Cochin', 'Delhi', 'Guwahati', 'Ghaziabad', 'Goa', 'Haldwani',\n", " 'Indore', 'Jammu', 'Nagpur', 'Parwanoo', 'Patna', 'Rishikesh',\n", " 'Rohtak', 'Vijaywada', 'Zirakpur'], dtype=object)" ] }, "execution_count": 140, "metadata": {}, "output_type": "execute_result" } ], "source": [ "matrix.source.unique()" ] }, { "cell_type": "code", "execution_count": 143, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'Ahemadabad', 'Bbsr', 'Gauwhati', 'Zirakhpur'}" ] }, "execution_count": 143, "metadata": {}, "output_type": "execute_result" } ], "source": [ "set(freight.source.unique()) - set(matrix.source.unique())" ] }, { "cell_type": "code", "execution_count": 144, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "set()" ] }, "execution_count": 144, "metadata": {}, "output_type": "execute_result" } ], "source": [ "set(freight.dest.unique()) - set(matrix.dest.unique())" ] }, { "cell_type": "code", "execution_count": 149, "metadata": {}, "outputs": [], "source": [ "cities = {\n", " 'Ahemadabad': 'Ahmedabad',\n", " 'Bbsr': 'Bhubaneshwar',\n", " 'Gauwhati': 'Guwahati',\n", " 'Zirakhpur': 'Zirakpur'\n", "}\n", "# freight.replace({\"source\": cities}).head()\n", "freight.replace({\"source\": cities}, inplace=True)" ] }, { "cell_type": "code", "execution_count": 150, "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", "
datesourcedestqtycartageremarksamountcomputed_amount
02008-07-01PunePune15.05case7575.0
12008-07-01AhmedabadJamnagar30.018case540540.0
22008-07-01JaipurBagru6.020case120120.0
32008-07-02RaipurRaipur23.05case104115.0
42008-07-02ZirakpurBanur1.01000consignment10001000.0
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount \\\n", "0 2008-07-01 Pune Pune 15.0 5 case 75 \n", "1 2008-07-01 Ahmedabad Jamnagar 30.0 18 case 540 \n", "2 2008-07-01 Jaipur Bagru 6.0 20 case 120 \n", "3 2008-07-02 Raipur Raipur 23.0 5 case 104 \n", "4 2008-07-02 Zirakpur Banur 1.0 1000 consignment 1000 \n", "\n", " computed_amount \n", "0 75.0 \n", "1 540.0 \n", "2 120.0 \n", "3 115.0 \n", "4 1000.0 " ] }, "execution_count": 150, "metadata": {}, "output_type": "execute_result" } ], "source": [ "freight.head()" ] }, { "cell_type": "code", "execution_count": 151, "metadata": {}, "outputs": [], "source": [ "df = pd.merge(freight, matrix, how='left', on=['source', 'dest'])" ] }, { "cell_type": "code", "execution_count": 152, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "date 0\n", "source 0\n", "dest 0\n", "qty 0\n", "cartage 0\n", "remarks 0\n", "amount 0\n", "computed_amount 0\n", "slon 7136\n", "slat 7136\n", "dlon 7136\n", "dlat 7136\n", "lane 7136\n", "distance 7136\n", "dtype: int64" ] }, "execution_count": 152, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.isnull().sum()" ] }, { "cell_type": "code", "execution_count": 154, "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", "
datesourcedestqtycartageremarksamountcomputed_amountslonslatdlondlatlanedistance
42008-07-02ZirakpurBanur1.01000consignment10001000.0NaNNaNNaNNaNNaNNaN
112008-07-02CochinMattanchery38.010case380380.0NaNNaNNaNNaNNaNNaN
312008-07-03GhaziabadNoida17.0250consignment2504250.0NaNNaNNaNNaNNaNNaN
732008-07-04CochinTripunithura6.010case6060.0NaNNaNNaNNaNNaNNaN
902008-07-04GoaTalegao11.021case231231.0NaNNaNNaNNaNNaNNaN
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount \\\n", "4 2008-07-02 Zirakpur Banur 1.0 1000 consignment 1000 \n", "11 2008-07-02 Cochin Mattanchery 38.0 10 case 380 \n", "31 2008-07-03 Ghaziabad Noida 17.0 250 consignment 250 \n", "73 2008-07-04 Cochin Tripunithura 6.0 10 case 60 \n", "90 2008-07-04 Goa Talegao 11.0 21 case 231 \n", "\n", " computed_amount slon slat dlon dlat lane distance \n", "4 1000.0 NaN NaN NaN NaN NaN NaN \n", "11 380.0 NaN NaN NaN NaN NaN NaN \n", "31 4250.0 NaN NaN NaN NaN NaN NaN \n", "73 60.0 NaN NaN NaN NaN NaN NaN \n", "90 231.0 NaN NaN NaN NaN NaN NaN " ] }, "execution_count": 154, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df.isnull().sum(axis=1) > 0].head()" ] }, { "cell_type": "code", "execution_count": 155, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.12835823942861685" ] }, "execution_count": 155, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[df.isnull().sum(axis=1) > 0].amount.sum() / df.amount.sum()" ] }, { "cell_type": "code", "execution_count": 166, "metadata": {}, "outputs": [], "source": [ "data = df.dropna()" ] }, { "cell_type": "code", "execution_count": 207, "metadata": {}, "outputs": [], "source": [ "data = data[data.remarks == \"case\"].copy()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Explore" ] }, { "cell_type": "code", "execution_count": 180, "metadata": {}, "outputs": [], "source": [ "#!pip install altair" ] }, { "cell_type": "code", "execution_count": 161, "metadata": {}, "outputs": [], "source": [ "import altair as alt\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline" ] }, { "cell_type": "code", "execution_count": 184, "metadata": {}, "outputs": [], "source": [ "plt.style.use(\"ggplot\")" ] }, { "cell_type": "code", "execution_count": 185, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 185, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAD8CAYAAACLrvgBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "data.distance.hist(bins=40)" ] }, { "cell_type": "code", "execution_count": 186, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAD8CAYAAACcjGjIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "data.cartage.hist(bins=40);" ] }, { "cell_type": "code", "execution_count": 183, "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", "
datesourcedestqtycartageremarksamountcomputed_amountslonslatdlondlatlanedistance
22702008-07-21KolkataAndaman63.0204case1282112852.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
28612008-07-25KolkataAndaman6.0204case12211224.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
63362008-08-18KolkataAndaman35.0204case71237140.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
63392008-08-18KolkataAndaman3.0204case611612.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
117372008-09-20KolkataAndaman60.0204case1221012240.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
176692008-10-30KolkataAndaman43.0204case87518772.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
245202008-12-08KolkataAndaman56.0204case1139611424.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
306502009-01-19KolkataAndaman59.0204case1200712036.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
351592009-02-09KolkataAndaman51.0204case1037910404.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
425292009-03-23KolkataAndaman59.0204case1200712036.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
462402009-04-14KolkataAndaman0.0204case00.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
464482009-04-14KolkataAndaman32.0204case65126528.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
509282009-05-07KolkataAndaman48.0204case97689792.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
580382009-06-18KolkataAndaman28.0204case56985712.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
589932009-06-24KolkataAndaman28.0204case56985712.088.3322.6392.6911.68KOLKATA to ANDAMAN483.33
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount \\\n", "2270 2008-07-21 Kolkata Andaman 63.0 204 case 12821 \n", "2861 2008-07-25 Kolkata Andaman 6.0 204 case 1221 \n", "6336 2008-08-18 Kolkata Andaman 35.0 204 case 7123 \n", "6339 2008-08-18 Kolkata Andaman 3.0 204 case 611 \n", "11737 2008-09-20 Kolkata Andaman 60.0 204 case 12210 \n", "17669 2008-10-30 Kolkata Andaman 43.0 204 case 8751 \n", "24520 2008-12-08 Kolkata Andaman 56.0 204 case 11396 \n", "30650 2009-01-19 Kolkata Andaman 59.0 204 case 12007 \n", "35159 2009-02-09 Kolkata Andaman 51.0 204 case 10379 \n", "42529 2009-03-23 Kolkata Andaman 59.0 204 case 12007 \n", "46240 2009-04-14 Kolkata Andaman 0.0 204 case 0 \n", "46448 2009-04-14 Kolkata Andaman 32.0 204 case 6512 \n", "50928 2009-05-07 Kolkata Andaman 48.0 204 case 9768 \n", "58038 2009-06-18 Kolkata Andaman 28.0 204 case 5698 \n", "58993 2009-06-24 Kolkata Andaman 28.0 204 case 5698 \n", "\n", " computed_amount slon slat dlon dlat lane \\\n", "2270 12852.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "2861 1224.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "6336 7140.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "6339 612.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "11737 12240.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "17669 8772.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "24520 11424.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "30650 12036.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "35159 10404.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "42529 12036.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "46240 0.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "46448 6528.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "50928 9792.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "58038 5712.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "58993 5712.0 88.33 22.63 92.69 11.68 KOLKATA to ANDAMAN \n", "\n", " distance \n", "2270 483.33 \n", "2861 483.33 \n", "6336 483.33 \n", "6339 483.33 \n", "11737 483.33 \n", "17669 483.33 \n", "24520 483.33 \n", "30650 483.33 \n", "35159 483.33 \n", "42529 483.33 \n", "46240 483.33 \n", "46448 483.33 \n", "50928 483.33 \n", "58038 483.33 \n", "58993 483.33 " ] }, "execution_count": 183, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data[data.cartage > 100]" ] }, { "cell_type": "code", "execution_count": 199, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAENCAYAAADOhVhvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/IgXRO8AAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "data.plot(kind=\"scatter\", x=\"distance\", y=\"cartage\", alpha=0.05);" ] }, { "cell_type": "code", "execution_count": 200, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAENCAYAAADOhVhvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "data.plot(kind=\"scatter\", x=\"distance\", y=\"cartage\", alpha=0.05, logy=True);" ] }, { "cell_type": "code", "execution_count": 213, "metadata": {}, "outputs": [], "source": [ "data.loc[data.distance==0, \"distance\"] = 5" ] }, { "cell_type": "code", "execution_count": 214, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAERCAYAAACKHYuuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "data.plot(kind=\"scatter\", x=\"distance\", y=\"cartage\", alpha=0.05, logy=True, logx=True);" ] }, { "cell_type": "code", "execution_count": 216, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAADxCAYAAAD2t6FzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAACs+AXqqixtqGEg4wAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "data.plot(kind=\"scatter\", \n", " x=\"distance\", \n", " y=\"cartage\", \n", " c=\"amount\",\n", " alpha=0.05, \n", " logy=True, \n", " logx=True,\n", " cmap='viridis');" ] }, { "cell_type": "code", "execution_count": 219, "metadata": {}, "outputs": [], "source": [ "data.to_csv(\"data.csv\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Model" ] }, { "cell_type": "code", "execution_count": 220, "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", "
datesourcedestqtycartageremarksamountcomputed_amountslonslatdlondlatlanedistance
02008-07-01PunePune15.05case7575.073.8518.5373.8518.53PUNE to PUNE5.00
12008-07-01AhmedabadJamnagar30.018case540540.072.6023.0370.0622.46AHMEDABAD to JAMNAGAR282.60
22008-07-01JaipurBagru6.020case120120.075.8126.9275.5326.82JAIPUR to BAGRU30.66
32008-07-02RaipurRaipur23.05case104115.081.6521.2381.6521.23RAIPUR to RAIPUR5.00
52008-07-02RaipurSaripali (C.G.)4.021case8484.081.6521.2382.5321.18RAIPUR to SARIPALI (C.G.)98.64
\n", "
" ], "text/plain": [ " date source dest qty cartage remarks amount \\\n", "0 2008-07-01 Pune Pune 15.0 5 case 75 \n", "1 2008-07-01 Ahmedabad Jamnagar 30.0 18 case 540 \n", "2 2008-07-01 Jaipur Bagru 6.0 20 case 120 \n", "3 2008-07-02 Raipur Raipur 23.0 5 case 104 \n", "5 2008-07-02 Raipur Saripali (C.G.) 4.0 21 case 84 \n", "\n", " computed_amount slon slat dlon dlat lane \\\n", "0 75.0 73.85 18.53 73.85 18.53 PUNE to PUNE \n", "1 540.0 72.60 23.03 70.06 22.46 AHMEDABAD to JAMNAGAR \n", "2 120.0 75.81 26.92 75.53 26.82 JAIPUR to BAGRU \n", "3 115.0 81.65 21.23 81.65 21.23 RAIPUR to RAIPUR \n", "5 84.0 81.65 21.23 82.53 21.18 RAIPUR to SARIPALI (C.G.) \n", "\n", " distance \n", "0 5.00 \n", "1 282.60 \n", "2 30.66 \n", "3 5.00 \n", "5 98.64 " ] }, "execution_count": 220, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Do we need to take date column in modeling? \n", "\n", "Let us see if date has any effect on the cartage." ] }, { "cell_type": "code", "execution_count": 226, "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", "
qtycartageamountcomputed_amountslonslatdlondlatdistance
count166.000000166.0166.000000166.0000001.660000e+021.660000e+021.660000e+021.660000e+021.660000e+02
mean49.60843426.01314.0240961289.8192777.759000e+011.298000e+017.514000e+011.535000e+012.800100e+02
std35.4851700.0939.946884922.6144101.140308e-131.425385e-149.977697e-144.632502e-141.140308e-13
min1.00000026.026.00000026.0000007.759000e+011.298000e+017.514000e+011.535000e+012.800100e+02
25%23.25000026.0615.750000604.5000007.759000e+011.298000e+017.514000e+011.535000e+012.800100e+02
50%42.00000026.01112.5000001092.0000007.759000e+011.298000e+017.514000e+011.535000e+012.800100e+02
75%71.75000026.01900.5000001865.5000007.759000e+011.298000e+017.514000e+011.535000e+012.800100e+02
max179.00000026.04741.0000004654.0000007.759000e+011.298000e+017.514000e+011.535000e+012.800100e+02
\n", "
" ], "text/plain": [ " qty cartage amount computed_amount slon \\\n", "count 166.000000 166.0 166.000000 166.000000 1.660000e+02 \n", "mean 49.608434 26.0 1314.024096 1289.819277 7.759000e+01 \n", "std 35.485170 0.0 939.946884 922.614410 1.140308e-13 \n", "min 1.000000 26.0 26.000000 26.000000 7.759000e+01 \n", "25% 23.250000 26.0 615.750000 604.500000 7.759000e+01 \n", "50% 42.000000 26.0 1112.500000 1092.000000 7.759000e+01 \n", "75% 71.750000 26.0 1900.500000 1865.500000 7.759000e+01 \n", "max 179.000000 26.0 4741.000000 4654.000000 7.759000e+01 \n", "\n", " slat dlon dlat distance \n", "count 1.660000e+02 1.660000e+02 1.660000e+02 1.660000e+02 \n", "mean 1.298000e+01 7.514000e+01 1.535000e+01 2.800100e+02 \n", "std 1.425385e-14 9.977697e-14 4.632502e-14 1.140308e-13 \n", "min 1.298000e+01 7.514000e+01 1.535000e+01 2.800100e+02 \n", "25% 1.298000e+01 7.514000e+01 1.535000e+01 2.800100e+02 \n", "50% 1.298000e+01 7.514000e+01 1.535000e+01 2.800100e+02 \n", "75% 1.298000e+01 7.514000e+01 1.535000e+01 2.800100e+02 \n", "max 1.298000e+01 7.514000e+01 1.535000e+01 2.800100e+02 " ] }, "execution_count": 226, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data[(data.source=='Bangalore') & (data.dest == 'Hubli')].describe()" ] }, { "cell_type": "code", "execution_count": 231, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.000000 1410\n", "2.462961 1\n", "0.773565 1\n", "2.386378 1\n", "3.829708 1\n", "4.712121 1\n", "3.601470 1\n", "1.054093 1\n", "0.450225 1\n", "18.961218 1\n", "2.905092 1\n", "4.284857 1\n", "1.388730 1\n", "7.807881 1\n", "6.966903 1\n", "4.569226 1\n", "2.745873 1\n", "0.472742 1\n", "1.190891 1\n", "0.876162 1\n", "1.426785 1\n", "0.551364 1\n", "Name: cartage, dtype: int64" ] }, "execution_count": 231, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.groupby(['source', 'dest']).std().cartage.value_counts()" ] }, { "cell_type": "code", "execution_count": 233, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYEAAAD8CAYAAACRkhiPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKgERERtTCYiI2JhKQETExlQCIiI29n+DKk0iN+JsVQAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "data.groupby(['source', 'dest']).std().cartage.hist();" ] }, { "cell_type": "code", "execution_count": 247, "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", "
sourcedestslatslondlatdloncartagedistanceqtyamountcount
0AhmedabadAhmadabad23.0372.623.0372.6018.05.0026.04683
1AhmedabadAhmedabad23.0372.623.0372.6018.05.0012236.0220248797
2AhmedabadAhmedabd23.0372.623.0372.6018.05.0033.05944
3AhmedabadAmreli23.0372.621.6071.2225.0161.25111.0277515
4AhmedabadAnand23.0372.622.5572.9522.042.40155.0341021
\n", "
" ], "text/plain": [ " source dest slat slon dlat dlon cartage distance \\\n", "0 Ahmedabad Ahmadabad 23.03 72.6 23.03 72.60 18.0 5.00 \n", "1 Ahmedabad Ahmedabad 23.03 72.6 23.03 72.60 18.0 5.00 \n", "2 Ahmedabad Ahmedabd 23.03 72.6 23.03 72.60 18.0 5.00 \n", "3 Ahmedabad Amreli 23.03 72.6 21.60 71.22 25.0 161.25 \n", "4 Ahmedabad Anand 23.03 72.6 22.55 72.95 22.0 42.40 \n", "\n", " qty amount count \n", "0 26.0 468 3 \n", "1 12236.0 220248 797 \n", "2 33.0 594 4 \n", "3 111.0 2775 15 \n", "4 155.0 3410 21 " ] }, "execution_count": 247, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# roll up\n", "\n", "agg_func = {\n", " \"cartage\": [\"mean\"], \n", " \"distance\": [\"mean\"], \n", " \"qty\": [\"sum\"], \n", " \"amount\": [\"sum\", \"count\"]\n", "}\n", "\n", "newdata = (data\n", " .groupby([\"source\", \"dest\", \"slat\", \"slon\", \"dlat\", \"dlon\"])\n", " .agg(agg_func)\n", " .reset_index())\n", "\n", "newdata.columns = newdata.columns.droplevel(1)\n", "newdata.columns = list(newdata.columns)[:-1] + ['count']\n", "newdata.head()" ] }, { "cell_type": "code", "execution_count": 252, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAENCAYAAADOhVhvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "newdata.plot(kind='scatter', x='distance', y='cartage', alpha=0.3);" ] }, { "cell_type": "code", "execution_count": 254, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAERCAYAAACKHYuuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCIIglEWMhSAIglAWMRaCIAhCWf5/I0FQxoWuxrMAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "newdata.plot(kind='scatter', x='distance', y='cartage', alpha=0.3, logy=True, logx=True);" ] }, { "cell_type": "code", "execution_count": 257, "metadata": {}, "outputs": [], "source": [ "newdata['distance_log'] = np.log10(newdata.distance)\n", "newdata['cartage_log'] = np.log10(newdata.cartage)\n", "X = newdata[['distance_log']]\n", "y = newdata.cartage_log" ] }, { "cell_type": "code", "execution_count": 258, "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", "
distance_log
00.698970
10.698970
20.698970
32.207500
41.627366
\n", "
" ], "text/plain": [ " distance_log\n", "0 0.698970\n", "1 0.698970\n", "2 0.698970\n", "3 2.207500\n", "4 1.627366" ] }, "execution_count": 258, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X.head()" ] }, { "cell_type": "code", "execution_count": 259, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 1.255273\n", "1 1.255273\n", "2 1.255273\n", "3 1.397940\n", "4 1.342423\n", "Name: cartage_log, dtype: float64" ] }, "execution_count": 259, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y.head()" ] }, { "cell_type": "code", "execution_count": 260, "metadata": {}, "outputs": [], "source": [ "from sklearn.linear_model import LinearRegression" ] }, { "cell_type": "code", "execution_count": 262, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)" ] }, "execution_count": 262, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model = LinearRegression()\n", "model.fit(X, y)" ] }, { "cell_type": "code", "execution_count": 263, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(1.1983538178610935, array([ 0.06244738]))" ] }, "execution_count": 263, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.intercept_, model.coef_" ] }, { "cell_type": "code", "execution_count": 295, "metadata": {}, "outputs": [], "source": [ "def show_predicitons(X, y, y_pred):\n", " plt.scatter(X, y)\n", " plt.plot(X, y_pred, color='b')" ] }, { "cell_type": "code", "execution_count": 297, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "y_pred = model.predict(X)\n", "show_predicitons(X, y, y_pred)" ] }, { "cell_type": "code", "execution_count": 300, "metadata": {}, "outputs": [], "source": [ "from sklearn.model_selection import cross_val_score" ] }, { "cell_type": "code", "execution_count": 322, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "-0.083758488565365427" ] }, "execution_count": 322, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model = LinearRegression()\n", "score = cross_val_score(model, \n", " X, y, \n", " scoring='neg_mean_squared_error', \n", " cv=5, \n", " n_jobs=-1,)\n", "np.mean(score)" ] }, { "cell_type": "code", "execution_count": 311, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([-0.06201186, -0.04173291, -0.3876596 , -0.02725678, -0.08762837])" ] }, "execution_count": 311, "metadata": {}, "output_type": "execute_result" } ], "source": [ "score" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Lets build another model with 2 features." ] }, { "cell_type": "code", "execution_count": 318, "metadata": {}, "outputs": [], "source": [ "X2 = newdata[['distance_log', 'count']]" ] }, { "cell_type": "code", "execution_count": 321, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "-0.084442034144584455" ] }, "execution_count": 321, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model = LinearRegression()\n", "score = cross_val_score(model, \n", " X2, y, \n", " scoring='neg_mean_squared_error', \n", " cv=5, \n", " n_jobs=-1,)\n", "np.mean(score)" ] }, { "cell_type": "code", "execution_count": 325, "metadata": {}, "outputs": [], "source": [ "newdata['cartage_per_distance'] = newdata['cartage'] / newdata['distance']\n", "newdata['cartage_per_distance_log'] = np.log10(newdata['cartage_per_distance'])" ] }, { "cell_type": "code", "execution_count": 324, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAERCAYAAAC6kZqPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBRFCCCExoyBCCCEkZv8fglMlEACwUbMAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "newdata.plot(kind='scatter', \n", " x='distance',\n", " y='cartage_per_distance',\n", " logx=True,\n", " logy=True);" ] }, { "cell_type": "code", "execution_count": 329, "metadata": {}, "outputs": [], "source": [ "model = LinearRegression()" ] }, { "cell_type": "code", "execution_count": 330, "metadata": {}, "outputs": [], "source": [ "X = newdata[['distance_log']]\n", "y = newdata['cartage_per_distance_log']" ] }, { "cell_type": "code", "execution_count": 331, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "-0.083758488565365469" ] }, "execution_count": 331, "metadata": {}, "output_type": "execute_result" } ], "source": [ "score = cross_val_score(model, \n", " X, y, \n", " scoring='neg_mean_squared_error', \n", " cv=5, \n", " n_jobs=-1,)\n", "np.mean(score)" ] }, { "cell_type": "code", "execution_count": 332, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAD8CAYAAABjAo9vAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJztnXt0VOW5/797ZnIn90kyJBDlogQkKJwAar3UI6Ut9Rw8\nrRUt1XJqlqs/BBHBVrkZQAQLOcjNHvsLilp+LtvayjktVcqyYhUFAoR7wkUwOEPu5ELIbfbevz/e\nzGQu+zaZ256Z57OWC/POntnP7Ey++53nfd7vw4miKIIgCIKIGgzhDoAgCIIILCTsBEEQUQYJO0EQ\nRJRBwk4QBBFlkLATBEFEGSTsBEEQUQYJO0EQRJRBwk4QBBFlkLATBEFEGSTsBEEQUYYpXCe22Wzh\nOrUTs9mMpqamcIehGYo3uFC8wSfSYtZbvPn5+ZqOoxk7QRBElEHCThAEEWWQsBMEQUQZJOwEQRBR\nBgk7QRBElEHCThAEEWWErdxxMAiNdcCunRBbW8BlZAEzZ8OQYwl3WARBELoiYoRdaKyDuHEF0FgH\nABAB4KsaCAtXkbgTBEG4EDmpmF07naLupH8GTxAEQQwQMcIutrb4NE4QBBGrRIywcxlZPo0TBEHE\nKhEj7Jg5G/DMpedY2DhBEAThJGIWTw05FggLV1FVDEEQhAoRI+wAE3eULgp3GARBELomclIxBEEQ\nhCZI2AmCIKIMEnaCIIgoI+KE/ciROLzySip4PtyREARB6JOIWjwFgJUr01FZGY/Nm1Mxe3Yn1q1r\ng0HntyfyuCEIIpToXBK9+f3vm3DPPd0AgJ07UzB8eD5WrEiDKIY5MBkcHjfigX1AzQmIB/ZB3LiC\niT1BEEQQiDhhT0gA3n23BefP2zB5cg8AYPv2IRg2LB9r16bqT+DJ44YgiBATccLuICkJ+OCDZpw9\newXjx/cCALZuTcWwYfl49dUhYY5uAPK4IQgi1ESssDtISRHx0UdNOHPmCkaN6gMArF+fhoKCfLz+\nekqYoyOPG4IgQk/EC7uDtDQRn37aiJMnryA/3w4AWLUqHQUF+XjrreTwBUYeNwRBhJioEXYHmZki\nDh1qwLFjdcjIEAAAS5ZkoKAgH++9lxTyeAw5FnALV4Gbei8wphjc1HvBUXMQgiCCSMSVO2rFbBZw\n6lQdrlwx4K678tDdzeHZZzPx7LOZeO21Fsyc2R2yWMjjhiCIUBK1wu5g6FABFy5cwTffGDF1ah4A\nYO7cLMydC/zhD324887QxOFLLTvVvRME4Q9RL+wOhg3jYbXacPGiEXfdxQT+xz+OA5CPnTub8e1v\n9wTt3L70a1U6FmZz0GIkCCJ6iLocuxojRjCB/8c/Gpxjs2dno6AgH/v3xwfnpDK17OLa5yBUlLtv\nVqK6d4Ig/CTmhN3BzTfb0dPTi48+GhD4H//YjIKCfFRWxgX0XLI16x1tXjtRqe6dIAh/iVlhdzB+\nvB1Wqw3/+7+NzrGZM3NQUJCPEycCI/CqNesuM3LZY5vqYa+zBSQegiCim5gXdgeTJvXBarXhT39q\nco5973tM4Kur/VyKkKpl90BsbWH59e4uwCRxQ2luQGvZAvKYIQhCFRJ2D6ZO7YXVasO77w4I/P33\n56KgIB/nzxsH9Zputeyp6dIHJSaxRdNjBwF7n+QhfL2Vcu0EQagSEGF/7bXXUFpaikWLoqdW+557\nmMDv2NHsHLv33jwUFOSjttZ3gTfkWGAoXQTuhfXSO1EB70VTCSjXThCEGgER9m9/+9tYsmRJIF5K\nd3znOz2wWm14/fUBQb3jjjwUFg6F1er75ZPbiYruLm0v0FRP6RiCIBQJiLCPGzcOQ4box1ExGDzw\nQDesVhs2b74KAOB5DlOmWDBmjAUNDb5dRsfs3bh4DQyli5jYazUFa24gP3eCIBShHLuP/OhHXbBa\nbfj1r1sBANeuGTBxogUTJ+ahpcWPyym1wJqVA6RLCP4g6tqFxjoIFeXgNyz1rp0nCCKq4EQxMK0p\nGhoa8Morr6C8vFzy8b1792Lv3r0AgHXr1qG3tzcQp/ULk8kEu93u12ts3WrAokUDVTMjRoj48ss+\nZGT4/lr2Ohs63/0t+JYmGLPMSHn0SbRvexl9J494HRs3fhKyVm/V/LqtZQvY4ms/xrwCZJRtgsmS\n73ugGgnE9Q0lFG/wibSY9RZvfLy2TZQhsxSYNm0apk2b5vy5qalJ4ejQYDab/Y7jkUfYf9u2DcHL\nL6fh4kUOeXns4v/zn/UYOdKHrtumeOCxeQAAAUArACElVfJQe0qq5tiFHVsguog6wCpsWnZsgSGI\n5mSBuL6hhOINPpEWs97izc/XNhGjVEyAeOqpa7BabfjFrIvOsbvvZlU0NluAUzQ++rnTblaCiC0C\nIuyvvvoqli1bBpvNhl/84hf4+OOPA/GyEQdffQJLOh/B198vcRufPNkyaIF3raKJGz9pUH7u1MWJ\nIGKLgKRinnnmmUC8TOSzYxMg8OA4oHZGCXjRgBF/O+h8ePJkJsZHj9YhN1fQ/LIOP/cssxkNZ04C\nu3aC98XSd+Zs4Owp4KrLV8pMM3VxIogoJWZse0PC9U63H42cgNoZJehLSMOoPw98i5k4kQnxmTNX\nkJamfe3aXmfTbP/rBccp/0wQRNRAOfZAkizdPDtuSCKsVhvOnLniNj527FAUFOSjS+PepM53fzs4\nS99dO4GWRvexlkayJyCIKIWEPZDMWQAYPOwGDEY2DtZw22q14ehRd3EePTofBQX5UKuq4lukV+fV\nFkFlF0+plp0gohJKxQQQY1Ex+IWrWK79eiebwc9ZAGNRsdtxubkCrFYbLl0y4lvfynOO33ADK2X6\n5hubZKbEmGWGlD0Yl5Gl2E6Py8iCZMLH+jWExjpqu0cQUQYJe4AxFhUD6yo0HXvjjayb08mTJnz3\nu7nO8WHDmMBbre7+6ymPPonuM8fd0zE5Foh3TQeUcu8zZwNHvwR6Pdr/9XRDXPEU+KRkYOQYcLNK\npVv1vb0VOH8GEAQgLQN44lmvmxVBEPqBUjE6wNHs4/333VMtBQUsRePAZMmXNBDjPtujnnuXy/PY\n+4CONuDYQYgblrpZDQiNdRDXLAaqj7PjBB5obQY2LgdffcLv900QRHAgYdcRt9/OrII3brzqNu4q\n8FIGYqobkHbtZKKshseCqvheBdDZ7n2cILB0E0EQuoSEXYc8/DAzGlu2rM1tPCEh3m0G70BtA5Iv\nO0zdjv2qRv5Aj9JOgiD0Awm7jvk//6cTVqsNDz543W3cM0WjZjvgyw5Tx7FCYx3QdV3+QJnSToIg\nwg8tnkYA27a1Ytu2Vjz0kAVffDFwL3aIu9UKCAtXyVbFYOZsNvtWK2/MygFmzma59Y0rZFv0AXCW\ncDrgq094VQPhrvtU35tSNQ9BEIODhD2C+OQTO5qamjBhQh6amwfq5ZnA58NqlRZEQ44FwsJVEMuX\nAc0N3geY4oBbJjqrYoSKcuWbwIyH3api+OoTrCrHkcfv6gQ2rkB3+mZg6HDZl3HeQAazk5YgCFko\nFROBHD9e71UKCUikaFww5FgAc57kYxhVBOO8ZU4xVc3JH/jE/ed+jxw3BB5taxYpN/TYtXNwO2kJ\nglCEhD2CsVptPgm8VpdH1Zy858Kp3EJqT7diGz+yEyaI4ECpmCjAIe6eYj6Qg7eBP7APOLzf+8lS\n3u5qOXnPhdPkFJZ+kcIxA5do6CG3I9b1xsJXnwAqyoFrbQBnAEaPA/f4U5SqIQgFaMYeRSjN4At/\n+Kj3YuioIklvd4cHPIomeJ/ExfvGyX88phiX7AxcpZqHrz4BlC8D2loAnmfxVx+D+Mrz1LOVIBSg\nGXsUIjeDL9xdCYB5xQMALtTIznwNORZg0UuS1S6edgLciUppLxrH4zKpHceirmxVzI5NgNQrt7VA\nXPschHG3URUNQUhAwh7FaBZ4BbR43yjmxD1SPZLljXJ9V5U2QXW0QTywD6g6AH7ecvKuIQgXSNhj\nAIeAOwTdQeHuSqDA22zMV+Ry5YZcC8RnBlI9Ppc3KuXuHfR0M++aG0aDyx1KM3iCAOXYYwMju3/X\nziiRnKUrlUlqQiZXnrlyK8TmRvDPl4J/+lGIZfMlyxvF8mXSOfM5CwBo6PQkCMDFsxAP7FOswiGI\nWIETRVF7b7YAYrP5N0sMBGazGU1N0s0r9Mhg4+VXLgC+ueg17jmDd6B1Bu+aVkFiEhvsaAPargJp\nGTANSYX91FEmvGr0L9h6zrbdqmIEAdDycU1NBzeI/HusfB7CSaTFrLd48/O1TcAoFRMLtEnnwGtn\n3Q/jf72jWCYph2daBQAQFw+IArMIbm6ASkMod2TKIo1FxcCGHQPnXPk0S78o4ci/H9kPftxESZ95\ngohmSNhjAV7Gsrd/XEsdvBdSu0b7ev0K03MR1tnk49xpgLezlNLQ4UCDzbtpiBR9fcxn3lZLNgVE\nTEE59lggPkHTuC87WYOxO9S1LFJorIO47pesyQffP/fn7SylFBcP3DoFGDkGSEhUf2GyKSBiDJqx\nxwKZZtb5yJMss+ThVqsNojjQos9BQUE+MjN5nDxZL99H1Y8Yxe4u8BuWstfu7gLaW6WP7ewAl5gE\nw7xlA3n+01Usvy/DYG5E5DxJRCok7DEAl2uBeNG7aQanIFIcxwS+83I9br59onP86lUjCgry8cB3\nluG1fA1WwFpISWUnPHYQQH8ZZFyc4lPE01XgX34OaGcLtRg5Brh8kXWBkno/PnjSA+Q8SUQ2lIqJ\nBWbOHqhacZCY5O0R44HQWIfE3yxH7YwSfHnfD9we+8vfM1H41l/whrCMiSqnoSxRjr5eb0HuU/CC\nB9js/GINsyG+eJbdFEQRKLqV2RC7IuWHowY5TxIRDAl7DCDu+QDo7nIf7O5i40q4iFt+Uj1qZ5Tg\n/dufcDuk7MMHUbh1Jz5L++kgo+PkF0I5Hz+eV5vApWeAW7XNq+G3r7Nscp4kIhkS9ljgk92+jfcj\nJWKTs46hduF/Yv169/z3T95dgMLdlbjUWeBjcAqZ+gklzIjMVeCNRvnjwWKWavjtK1otjglCj5Cw\nE7IoidtPfnIdVqsNjzzivuX/nn27ULi7Etfsyf6dPMcCblYpuMfnAebcgXG50k2X2ITGOggV5eA3\nLIVQUe7TTlTHc8WGOu+Km8GkdAgiDNDiKSGPlC+7h7iVl7ehvLwNd9yRi9ragY/TuD2fAgAufX8y\nDAYAaZnsAdfNUgmJ0puNsnOd6RPVNn2u5Fgg3jWdtekbxKKnvc7mvenKYADyC8EV3EBVMUTEQDP2\nWCBnqG/j/Th82bXkq7/4ogGXq454jd/4t0Mo/OshJugGA6s/738tzFsu6THDLXpJvU2fKY4tAMcn\nAMNudMbGfbZn0Iuene/+1vu5gsDGSNSJCIJm7LHAkDSg8Yr3eGqa6lMNORbJ7kdyx9a+MBc4dlDa\nSRLA5ZWLYJi3zDkuLFwF8b0K9s0AAPIL2bhDjG210iez9w00Dunpdgov78eiJ98i4wnS0y3bBYog\n9AgJeyzQflV6vE1m3B/6q2/krIKHv1gOvOhhU2CrHdhcdOwgxBOVrHxSJZ/uxMVnRku7PTmMWWbI\nFVk6bgy0aYmIBEjYY4G0DFbvLTWOwIqVp7DKCbzDouDySokcupwbZGo6+1dih6lzRi61LhAXB7G7\nC0JjneL7Snn0SXR/uU+6/DIxSXrT0uH94G8hozFCX5CwxwJy9eCcQXaHJf/4fHCf7fFd7GUaYdfO\nKAFyLCh86y9u48NfLB94XA1zHnBVwhoBcG7AMuRYwM+cDezYPJCqcZiBVR9X7LZksuSzzVbVx6XP\nIbVpyU5GY4T+IGGPBS5/JT8ut8Ny62qI/RUrvlSWOPqYiu9VgPuqBmLXdVb9MnosuFmlsL7sR7s+\nW62qZS9ffQJ4Y6P0rL+nG9i8EsLKre5dnfpz/A0GA7MclqK7i/nXyCFjO0wQ4YCEPRaQEyu7XX5R\n0VNAfRUuWy1ER8rE3ue1CGq12iBUlDtn7A5kBZ7jlEW9jr0eDu9XbuzR18usgBe9xER9/RLgKls0\n1dKQW+kY0VYL/vlSt8bfXHYO5eSJkEPCHgsYDAAvIXYGg08ujWJri7Z8vJLPisuNQWxtUe7HCheB\nV+ucVP8NxPpvtL2RmhMDVTdXNXTHca3dl0gzObl8CU7p7+oENi6HOCTN6VJJRmJEqKA69lggSWYX\naFIyE6xMafteLxKTIG5cwboT1ZyQ7TGq1WfFtVJFrh9r4e5K2RZ+g0YUIZbNh1h1QP3YhETg8fnM\npqC/rh+3TvF2nzQY4DWfFwRv62EyEiNCQECEvaqqCgsWLMD8+fPxwQcqxlJE6FHooGTIsQCFI9Vf\nwzHD1LD5R7PPikQT7JAJfG+Peos9AOjpZpueXOASk4BhI4DsXGDkGLbZKi5e86nJSIwINn4LuyAI\n2L59O5YsWYKNGzfi888/xzffaPxKTIQGtQ5KcouCqeluO07ljvMSKgnBdmz3d/VwAcBe11HG6EJI\nZ/AquNawO7+xXDzLSkg72tj7HaK+2csBGYkRwcbvHPv58+dhsViQl5cHALjzzjtx6NAhDBs2zO/g\niAChIuyym3rG3QaDS05c0Lj5x1EZk/DhH9Fdf4W9/l3Tgbe3QPQoq+QWrgI37jYmlhJozsEHE1st\nMwbr7pL8xiKWLwPu/3fg9xXez01JBTo7Bn4mIzEiBPgt7C0tLcjOznb+nJ2djXPnzvn7skQgkbIT\ncB3XYPbl03Fg4p6+sAx9TWxxUqgod4r6wPn7Sw01EFaB72hjNx65rk7NDcCu30k/NnosuMQkqooh\nQkrIqmL27t2LvXv3AgDWrVsHs1njgl0QMZlMuohDK4ONt17hMbPZDJjNsK/ais53fwu+pQnGLDNS\nHn2SbdhxP1jbcRLxtnR2SG/XP13FOiipYMi1wJibj7rJz8B+8RyGvf1Xt8dDIvBKXZ1k8vVxvB1Z\nz68NSjiR9vkFIi/mSIvXgd/CnpWVhebmgd2Azc3NyMryziFOmzYN06ZNc/7c1KShzCzImM1mXcSh\nlUHHy3HS5YIcN/B6pnjgsXkAAAFAKwBInUvrcR7xCimp0rFpEHXkWCA+swqCw/GxsQ6XRy2CeLwS\nhe+7p3BCmqLRgD0lVfZ35q+VQ6R9foHIi1lv8ebnS0+iPPFb2EeNGoUrV66goaEBWVlZ2L9/P55+\n+ml/X5YIJN//MbD799LjoULGw0VyFpyQyBYj0zNZw20PwXM4TgoV5ajt0kEOXgHx4D/Bt10F9/g8\nt/cgaeVwYB/4uHjAUgAuv5DSNsSg8VvYjUYjfv7zn2PNmjUQBAH33Xcfhg8fHojYiEDhsMTVOh4E\nHAuqrjNUsbuLNaH2gLttqtuiLSA9u8XM2cCZY0B7qz4WWaUQBaD6OMQ1iyEs3TAg1FKbuAD2Deby\nRYiXLwLnToMfPgLo7qL8POETAcmxT5o0CZMmTQrESxHB4Pxp38aDhKe3u9BYB9FWq7oYK2dUxi1c\nBYy42e3moFuB72yH+PZWCI/PYzeo4xpKNlsa2X+gXauEb5ClQCwgyninyI2HCKlZvM8WBTK19boU\n+JoTEDcsdYq1z5DRGKEREvZYYEi6e69R1/Ewo6VDk5JFgZrXTe2MEtgFI0Z+6G4fEBaBF8XBi7rj\nJSo/B99QBy7XAvuc+WwxmyA8IK+YWKB0EQDOY5CLiJmf0FgHNEkXbDpz7SqpCZOBR+2MEhy+f7rX\nY4W7K/Hdf/6/gMQaEng7cLEG4oF9aC1b4OXTQxAACXtMYCwqBh5+ot+oCuzfh5+QbTihF5y5danu\nTy7VMq4Nt3HrFGfDbMS577jNSWBuku/f/oTb+JmOm1G4uxKv1MwN5tsJOHy9VfMGLyK2oFRMDMBX\nnwD++OaAT7kgAH98E/zwEfoWd7nKkexccC6LiFLpHKGxDuJK6bLbyVnHUDujBG9cmoWy0885x7dd\n+Dm2Xfg5tv/LQnwn75+Bex/B5NTRgVn7rp0QG64wR8m0THC53qWiRGxAM/ZYYMcmQPBweBR4Nq5j\nZF0QzXnqYrVrp6p7489vfA+1M0rwnVz3TU5PHN6Iwt2VuNRZ4Eu44cHeB/G9Cm9zsv50jZStMhH9\nkLDHAtc7fRvXCZrtfyVQtcblBtYctpcsklxEvWffLhTurkQXL2OipheUmn+Q/3tMQsIeCySn+Dau\nF+Tsf4tLwD9fCv7pR9m/1Se8nqoq/mO8U1ByVsFjPvochbsrVZs4hQ05v/1+xJNHaNYeY5CwxwJz\nFgAGo/uYwcjGdYznwig39V4m9m+8ytINXZ3s340rvMVdqTNUphn4wSy3WbsrcgJ/w99C7wWvSlw8\n6ymrRGcHxCVPgv/7/4QmJiLskLDHAMaiYmDhKtbxJymF/btwlb4XTvsx5FhgKF0E4+I1zGbgz+9o\nWi9Q7AxVOJJ1RVKZguup2YcUXGISMGos6walhd9XgJfxvSeiC6qKiRGMRcXAuigojfNlvaC9TfrY\njjbmU6MR3e1iNZmA0eMQl5qO3pNHfHtuRTn4z/YCCQkR5UHjrxNmrEHCHiNEzR9GcgpLwXjS18va\n7bm+r/ar0q/RdhXc6LGKO1al0I3A2+3AVzXo1TpT96T6mPN/I8GDRs4rSM8xhxtKxcQAbr06a05E\ndhmc1HoBwMr+PN9XWob0a6RlsBx8Vs6gQtBFimawoi5Ff3s/3X4elLyCCElI2GOBKPrDcFsvkBJ4\nl/fF5Q6VfA0udyib6Q0f4dvJM81AxkAbSF0IfKBobtDtzV7JK4iQhoQ9Boi2PwxjUTGM6yqAm8ZJ\nPu58XzLlkk5bYB/y7EhIBH70M+bV4kHUCLxOb/b+7GeIVSjHHgPIOSBG+h+G2vtSswVWc4Z0o6cb\nqChXPEQ3OXg/0OXN3ocm6gSDhD0WiNY/DA3vS9EWWOr5ASCiBb6pHkJjna4WJTX79hNOOFEMz346\nm80WjtO6obdGtWr4E284qmJCcX39fV989Qlg62pVXxl/kEvH6FbgExKBecuDss8hlv7mgoHWZtYk\n7Dr6palB8QYeoaKcVQuFgIgT+KJbAVEI6EQgEj4TrugtXq3CTqkYIqYJZU454lI0/fXuIgCcOw1h\n8RpKf0QIJOxETOPTAmqAiDiBB4CWRtbUY94yAFG04S1KIWEnYhupBVRTHDAkDWht9j6e41Q9ZrQS\ncQJ/7jQA2gkaCZCwEzGNa8WFqbMD9pRUZ1WNq3gBYBU3Q9JYM4sAEjEC79jtqrThrXQRzeZ1AAk7\nEfM4SiKzPBbKpErssGsnxAALuwPdC3xcPACFDW+nq1iV0dtbaDYfZkjYiYDhOVOzz5kPmOLDHdag\n8ayBFxrrmCukKU7dA90PdCvwCYkAFNYlOtqkS0ddZvNEaCBhJwKCVN619dJ5CAtejIqZmuf7CwW6\nE/ghaexfpY1dMvsBdLmjNYohrxgiMEjkXfl6qy69RwaFVF45RNTOKMHX39eBF803F8Ef2OfsbIXU\ndM1PjXT7ikiDZuxEQIg2ozFPZN9HXBzQJ5OWGTkG+OZSQCx2OY4JfKc9CWP3/NPtsZDO4CvKwf/1\n9+w9qfRadRIN9hURBs3YiYAQ7Q58su9j3ET5mWtcPIzb/sB6tQaIFFMXameU4O93z/J6rHB3Je78\nRwj6ml65zHrNXr+mfmxqOjhaOA05JOxEYJCwyDXmFUTPTE3GApibVQpu3G2ST3HcDILxrWVM6gXU\nzijBhuKVbuPfdOWjcHclVp95JuDnVMQUJznMjbvNTdS7TxwF/3wp+KcfZf96NiEnAoKxrKysLBwn\n7ujoCMdp3UhOTsb169fDHYZm9BwvlzIEmDAZ3LV2YEgauNFjkfnMi+iRmc0KjXUQ330dwsd/BU5X\nQRw+kr1GGFG6vlLvj5uzAIYcC8ThI4Hjh9xnsDkWcHMWsOedrgKsXwcl5lvSz2LhTb/F2Y4ROHdt\nlHP8SOsEbDz3JCakn8LIlMtBObcbggAYDO6bt1yvAZjhWs8rvwQ6r7Gqoq5O4MA+iDfdAoM5L/gx\nDgK9/c2lpqZqOo5MwHRk8KNGtMQrWWHSvyAXzq/swXLPDGVFjdxi6hf3/QAFSfVBPz/AAcNvBJdf\n6LUxiX++lKVwPMnOZY1T+tHTBie9/c2RCRihX1R2LkYiSr7vjt2t4trnWK13EJErkbzjH38FAFz4\n3lTEGTQueg4KEai3QUweAm7XTgj9oiw01klbNADA9YHm5GRXEBgox06EnGivoJHCkGORzcUHA7l2\nfaM+PBD8EsneHrem6Xz1CSbWclU0ySkD/x9F/XnDCc3YiZATra36VFHa2JOQGJRmH2Hf5NRYB2z/\nL/nZusEI/MdjzBe/tQWw1UoeFs03/WBAwk6Enghr1ReonK9bi7fGOqDtKpCWAS53KMS7prt5rASa\nsAq8nKhzBuCh/wTe2QZR5abmy01fTzn6cEGLpzpaGFEjmuLV4x+fVLyhXOh1XpOqg0BPV0Bf2xNd\ndHPiDIDRqO6748P1DvTvS29/c9QaTwN6+6WpQfEGF0lhl2mdx029F4YgLfTyG5YCNaGp79aFwEuR\nmg7kF/p80w/070tvn2GqiiGIABCOhd5QdnUKew5eBm7cbapCLPWtLxYX5qXwS9i/+OIL/OEPf4DV\nasXLL7+MUaNGqT+JICKIsCz0Sq1BZJqBrutAd3A2y+hK4DlOdb1FriwS+YXST2iqh9BYF/Z0X6jw\nq9xx+PDhWLx4McaOHRuoeAhCX8hYCQRzodfhnshNvRcYUwzcOgUoHMkMx4KMXJlkSJ0kR9ysLsBy\nZZE9Pd6/LwBoboC4cQWb5ccAfgn7sGHDNOd8CCIS8RRZbuq9Idkha8ixwFC6CNzP5rMSwGMHpTc3\nBamRSVgFvrVFVYBlUys1x9lNNzvX+7EYqoenHDsRNQSr0kZpV2nQkfOBT01HfFExei/UAC2NQTt9\nWFI0LY0Q36sA5i2TPUR2HUIUgXe2sZSMhH1BrOTaVYV99erVaG1t9Rp/5JFHMHnyZM0n2rt3L/bu\n3QsAWLduHcxmsw9hBgeTyaSLOLRC8cpjr7OhddNK1twDLOdqvHQeGWWbYLJorCTQ4fVt6eyAVDFg\n3A2jwHGC6oWGAAAWo0lEQVRcUEXdlZAL/OkqZNh7ZX939jnz0XzkC6Cv1/vBnm4YOtshSDwvMW8o\n0n34HevxM6GFgJQ7lpWV4bHHHvNp8ZTKHX2H4pVHrswNqelsK7+G2bser6/s+7p1CnONlBK2EBCS\nMklTHJCUDIwcA25Wqdfvj9/6EktRSTFyDEtd+VnPrrfPhNbUN3nFEFGB7FfsjjanZ0lELpzJLd4C\nYRN1IEQ5eHsfE+djByGuX+L9+5s2k1kFS8CFaW1EL/g1Yz948CDeeOMNtLe3IyUlBTfeeCOWLl2q\n6bk0Y/cdilce2ZmtC2qbVPR6fSXrtd/aErJNTFoIxQyem3ovu9Ht2gmx4QpbVJayIgjgzmC9fSZC\nskFpypQpmDJlij8vQRBO/Fr8VDLY6scxq9ejnYESUou3gtziYUoquPGTmPfMB+8AF6pDEmMocvBi\nYx2g5mufnuUm6pH2uw4UVBVD6AJ/fbjdDLZOV0mWBnIZWYrngQ+LZGEXjJmzYbx03rlYDMBrpiqY\n8yBKCXt8AkthCEJAGm27ElSBb7sq3ajDlesDndli2dudcuyEPgiAD7ez9vuF9fKbigJwHodgiAf2\nufmOhzKHb8ixIKNsk2IOWXbdYcTN4FZsAjfxdtlepf4S8Bw8x6mLOgD09UFc+xxLzb1XEbPe7jRj\nJ3RBID0+3GbvHjNqPhDn0UkHKJMlX3HNQLbWOzEpZK36AjaD92UpsH/BXO6mJTo2QEVxioaEndAF\ngfZkkdtUFIjzRIzRlJzvPSAt6gYjkJnNOhpdvhjQUMKy0UnODtj6NcQNS517AEQAqDoAft5yGIuK\nAx9HGCBhJ/RBqJpvaDgPX32Cdf1pb2W56NFjwT0+zzmji5QOUHLfXMS3tkg/4aZxMC5ew+rDAyzs\nDmpnlEAQOdz4t0Nu44W7K5Fi7MSZ794blPO6ca3de6ynG9i4HPwNN4HLtUT8DJ6EndAFSumTUJ6H\nrz4BbFzOFhYBQOCB6uMQ1/0KwvOvsOMiqAOULxU1oboxGTgRtTNK0N6XgvF/HyhR7eRTULi7Ev82\ndA+2TVwSkljcEATgYg3EizXA4f3gRRH1EIHUDOCJZyNqNk+NNnRUo6oGxRtczGYz6ksflF2kc62D\n10OOdrDXV63LUCgbfQDAV9cK8e1P/+Q1vnb8Gswu/HPI4lDEYAAWrg67uFMHJQ1EovBQvMHDbDaj\n/iffAbo6pQ8YOQZcjkU3C27+XF+lG5OWzV7B4OOGb2FO5Sav8T/e/gSmZB0LeTxeZOfCuK4irCFQ\nByWCGAzJKfLCbv0a4lc1APoX3A5+Cn5MsVv+PVJQdKyUSjXFxbP+pN3B68X6r7mfo3ZGCd64NAtl\np59zjj/05XYAwJf3/QD5SfVBO78q170/F3r45iYFzdgjbEZJ8QYPs9mM+s/+4Z5jd2CKk6+yiIsH\nho0I+aJbMK+vlGBh186QzuSfOvoy/vfKdK/xs9/9FhKNgd1YpQmPGXsoG507IBMwghgExqJiYOFq\nICOblf+Z4oCiCcDwEfJP6utli26RbDbmgWOzl3HxGhhKF8HQn4IKJdsmLkHtjBJkx7uf9+aPPkfh\n7kqfStsDQo7F/XcbgM1uwYKEnSA8MBYVw7j+TRhf/zOMv3kfxkUvgcsdqu3JOvnDDgbhKuc8Om26\nZJ37DX8LYbs+gFVHudy49byfgYSdILQwczaQkKjpUD38YQcFOQvhRWuAlNSgn37QNgWp6cy/PivH\n+7GkFLbWoDV14nLjlrvROcaFxjoIFeXgNyyFUFEeWsuJkJ2JICIYQ44FmLecGWipoLeNSoFCrv+r\nsagYGB26hvaDEvjuLiB3qPfvb0gquJFjwC1cJd0nVQLnjVuh0Xm4/YSoKoYgNGIsKoZQtoWZS31V\nM+CO6NrwQmKjkl4rJwZDWPu/eqDZpqCjTb4uv/93YyhdBGHRS9o8dBKTAChvdhMqysPqJ0TCThA+\nwGbuA02W1URb0jo2ynxJADD7hTDhrw+NYwbuKdSoswJt8mk1pd99uPPvJOwE4QeqM1ipyomebqB8\nKfi4eOBn82GcGgJ/lGAjJ+xJKaxMNARt/AYt8P0zcMD998lvWCot7N1dql7v4fYTohw7QQQRxRla\nXy9QUQ5+3S8jv0QyLVN6fOgwGF/7I8vLa0Gmh6kv+JyDr/1K8vorLo6qlToq5N9DAQk7QQQRTTO0\nC9URX//O5UqvGXAqqQkvPDeG+YFmgb/aJF2iOnM2jHkF7mP94qyWapFbaA7V2gqlYggimMycDVQd\nkG667EoYGnUEFBXHS9mmHyFAS4pGSqgdXapadmzxXhzVkGoJ50IzCTtBBBFDjgX8vOXA1tWq4h7J\n9e+qtsszZwOH90vbMiSlACaTZJ/aQFI7owQwGFD4l4Nu44W7K4HdgHWxzXtBdM586S5VOrduJmEn\niCBjLCqG8OJmiBXlTAxkiPT6d6UZqiHHAv6WicCxg16PcRPYjNonH5rsXMCcx74JdHdJvi4AINMM\nFI4EurvYsXdNR232A0BjndcMvqAgH0A+amc8yeIB0HzkC2DcbeBmlbqlUULVP2CwkLATRAgw5FiA\nF9aDP7AP2LHZe+aqo9lesOBmlcJQZwVfbx0YdH3fnjNgJdIymFC3trCqlkwzy5UPnA0YkgrkFbiJ\nOvfZHohD0gBBQO3P/g1ovKJcRdPXCxw7CNFW66x4caCnmn5PyN0xwtwHKd7gEcp4fdm0JHdspF1f\nAMiw90rmrAH394nEJODMMbYBTIqERPfUVnwCM23r6ZJvfG0wKC7Oyu1adeToXRuthAtqtKGBSPvD\noHiDix7jlbSGBYCMbKQ/uwrXhg4PT2CDxJdrzFefALas8hb3+AR5wQ8AUgJvAI9LC0thXLwmaOfV\nAtn2EkQ0IFUvDQCtzWhb+TQTvyjFWFQMrmwLM/BKTR8w8yq4MajnlSqTFGBE4cY38dRTGUE9d6Cg\nHDtB6BjFShmeBzYuB5+QxDo/zVkQXTYF8LZwAPpb912UX4QOFFJlkh98kIwPPkjGU091YMmSjqDH\nMFhoxk4QOka1UkYQWCu/5gZmU/D3/wlNYHLhhMKqVmpXZzBIz0Lc+Em4vHIRLlcdcXto27ZUFBTk\nY/v2lODHMQgox66znKoSFG9w0WO8sjl2JRatCcvMXUuruEBdY+dCa2MdYP3afSE1IRHILwTS+tMm\n3V1sMfbiOaD9qvaTZOUge81v0GqKdxtmZZHufPRRA8aPt6vH62dpJC2eakCPf8hKULzBRa/xCo11\nEN/eClQf1/aE+ARgxM0hr60WKsola9Fdq0mCcY21iqbQWAexfBn7dqORxHumo++xeZKPSQn8oUN1\nyM93r7wJZG9UrcJOOXaC0DmGHAuw6CW2ULpjE3C9k5X1yZXu9faw5g6Am+NgsAmXVa3WenKxuRFo\n82HGDoBvkb8JWa02iCIwbNiA2E6ezK7z8eN1yM7u//0oGYYFqXyScuwEESEYi4phXFcB4+Z3WcNt\nLYSwB6taq7hwwlefADaukLY0UMCYZVZ8nOOYwNfWumcgJkywoKAgH+3tXFhueCTsBBEm/FloNBYV\nI+k/F2g6NmQeNGG2qlVkxyZA4H17To4FKY8+qelQo5EJ/KVL7gI/duxQFG58E72Cd3IkmDc8SsUQ\nRBhQa9SghbR/n4WurNyB9Axvl9y4E7LmDnr2T7neqe04Fw8azJwNkyUf8GFNIC6OCXxXFzB69ECK\nZvSHX+JHBX9B+YSVMHBi0G94JOwEEQ5k8q7i2ucgjLtNsyAai4qBdRUA5BfpQjlj1q1/SnIKKwtV\nw5wXkN2lSUlM4Ds6OBQVDQUAvG99AP96ay3+feqFoN/wSNgJIgzIpkc62lhlyZH94MdN9HIVVMLX\nGXM0NdlWZc4ClmNXSccE+ttNaqoIq9WGnh7gz39Own3f/w8Y0oNfiEjCThBhQLXxRF+frKugElpn\nzJKpoMP7wd/i280kUjAWFYNfuArYVCa/gBrEbzcJCcAjj3QF5bWloMVTgggHWndPBquqRSoVZO+/\nmUR4mz45jEXF4P7lTukHs3ND2rou2JCwE0QYcOuJmZqueGwwqloUXzOEJZIhR6Zyh1v0UtSIOuBn\nKuadd97B4cOHYTKZkJeXh7lz5yIlRZ/eCQShNxxpEzXbgGBUtailgsSGK+C3vjTQ8WnkGJ9SNHrN\n34e6cidc18EvS4Fjx45h/PjxMBqN+N3vfgcA+OlPf6rpuWQp4DsUb3AJZ7xCYx3E9yqAU0fdc8AK\nW8/9iVfVg0bK89xgBIr/RVXgJV/bFAfcMhHZv/ill/dKIAiWgAb8Gg/SSsCBVksBY1lZWdmgzgDA\nYrHAYGDZnO7ubpw7dw5Tp07V9NyOjvBbXiYnJ+P69evhDkMzFG9wCWe8XMoQGKbcA0y9F9y1dmBI\nGrjRY8HNWSArAvHtrbj+xqsQPv4rcLoK4vCR4FKGaD4fJkwGmuqBlgZ3e4KEROlGFqII1FuB44eA\nCZNlzyW++zpw9pT7oCAA9Vb0Vn4OobhEc5xacAro2VPMB8b6tWqMWvHnMyF5Ha5fA3etHdwkmVy/\nCqmpqZqOC1hVzMcff4w77xxcsARBMHypamndtBJif//QwWxwcnide852xYY6QMnvXMXnRCl/z9db\nwQXaIyUMXixaCJd3DqBB2FevXo3W1lav8UceeQSTJ08GAPzpT3+C0WjE3XffLfs6e/fuxd69ewEA\n69atg9ms7MEQCkwmky7i0ArFG1wiKd62d7ai27UpNAA01iHhwz8ifWGZ4nPtdTZ0vvtb8C1NMGaZ\nkfLokzA9v3bgtTeWoVulkYWpswNZMteqLW8oumvkOzuZOjuQZu/1jsGiLc3gSUtnB6QKGJVi1Io/\nnwm565CYNxTpQf6cqQr78uXLFR//5JNPcPjwYaxYsQIcx8keN23aNEybNs35sx5yr5QDDi4Ub/Dg\n669IjnfXX0GfwnvwzPv2Aeg+c9wt7yt87yHg5FGgpVH2dewpqWg4c1Iyry187yHgzHHZ/H2f0YTm\nFfMUY/AFIUU6PWFPSfX79+lXjl3qOuRY0PO9hwb9miHpeVpVVYVdu3bhV7/6FRISEvx5KYIgfGDQ\nTopKaYt+DDkWcIvXsP6iKanMwtCVHAvEu6ZD3LiC7ZKtOQHxwD5n/bujlBO3TmHmKS4Y8woGzqkQ\ng0+EyXxMzcTNraR1TDG4qfeGrFberxz79u3bYbfbsXo1sxC96aab8OST2tzQCILwg5mzYbx0Hrxr\nOkZGzFxz6LDVSr6cZ97XtdeoVMWJs3uRK411EJc8Cd4UB4weC+7x/gYVLs/NmDMfza+u1BSDVsJh\nPqbVxC1c3jl+CfuWLVsCFQdBED5gyLEgo2wTWnZsURQzra31lGb6UuLEK4mwvQ+oPg5xxVPAgjIY\nXZ5rMptla+j9qdcPuYDqdMHWAXnFEESEYrLkO1vOySIlQJ5kmn1OW6h63QBM4LeuhvDiZvcbzszZ\nbONTGF0o/SWcFS9aIEsBgohiNAmNQtGDLFq9bnq6vXLn4cw9Bwo9d4sCaMZOEFGNppl1S6PPKQS3\nvPbh/Yot58TWFmeevqWzg1WxzJyt/m1Dz+j8WwcJO0FEM1ICJMFgUgiOvDZ/13Rg43L55tqJSc48\nv1P+Q9hkOxjoulsUSNgJIqrxFCA01bNt9x44UgiD8VxhXuergf+7Hmj32MzoeK6OFxoHi267RYGE\nnSCiHlcBUmqf508fVmNRMVD+tuSNQXxLunpOLwuN0QgJO0HEEEopBKGi3O+ZtdQsVghCeSOhDAk7\nQcQYcikE2RK+hisQKsoHn0vW+UJjNELCThAEAIUKGlstxItnAfSnZ3xstO36LcHU2QG7oypGJwuN\n0QgJO0EQDKmZdUIiq0V3ZRCNth3fErIiyGgtkqENSgRBAJDeOIT8G+SfEM29USMcmrETBOHEM/8u\nVJRDVPBmp8oWfULCThCEPCobnFwrW/TawDoWIWEnCEIWx8KnXKNtR2WLPzXwROAhYScIQhG53qhu\nM3Kd29jGGiTsBEFoQmkLvd5tbGMNEnaCIPwmGM0z9I6e1xRI2AmC8J8Y212q9zUFqmMnCMJvoqF5\nhk9oaAoeTmjGThBEQNCzjW2g0fuaAs3YCYIgfETvrfFI2AmCIHxFquerjtYUKBVDEAThI9QajyAI\nIgrR85oCpWIIgiCiDBJ2giCIKIOEnSAIIsogYScIgogySNgJgiCiDBJ2giCIKIMTRVGyMTlBEAQR\nmcT0jP35558Pdwg+QfEGF4o3+ERazJEWr4OYFnaCIIhohISdIAgiyjCWlZWVhTuIcDJy5Mhwh+AT\nFG9woXiDT6TFHGnxArR4ShAEEXVQKoYgCCLKiHp3x6qqKrz55psQBAH3338/HnzwQbfHRVHEm2++\niaNHjyIhIQFz584N61cvtXhPnTqFX//618jNzQUATJ06FQ899FA4QgUAvPbaazhy5AjS09NRXl7u\n9bjerq9avHq7vk1NTdi2bRtaW1vBcRymTZuGGTNmuB2jp2usJV49XePe3l68+OKLsNvt4Hket99+\nOx5++GG3Y/R0fTUjRjE8z4vz5s0T6+rqxL6+PnHx4sXi5cuX3Y45fPiwuGbNGlEQBLGmpkZ84YUX\nwhSttnhPnjwprl27NkwRenPq1CnxwoUL4rPPPiv5uJ6uryiqx6u369vS0iJeuHBBFEVRvH79uvj0\n00/r+jOsJV49XWNBEMSuri5RFEWxr69PfOGFF8Samhq3Y/R0fbUS1amY8+fPw2KxIC8vDyaTCXfe\neScOHTrkdkxlZSXuuececByHm2++GZ2dnbh69apu49Ub48aNw5AhQ2Qf19P1BdTj1RuZmZnO2WFS\nUhIKCgrQ0uLeV1NP11hLvHqC4zgkJiYCAHieB8/z4DjO7Rg9XV+tRLWwt7S0IDs72/lzdna214es\npaUFZrNZ8ZhQoSVeAKipqcHixYvx8ssv4/Lly6EM0Wf0dH21otfr29DQgIsXL2L06NFu43q9xnLx\nAvq6xoIg4LnnnkNpaSmKi4tx0003uT2u1+urRNTn2KONESNG4De/+Q0SExNx5MgRrF+/Hps3bw53\nWFGDXq9vd3c3ysvLMWfOHCQnJ4c7HFWU4tXbNTYYDFi/fj06OzuxYcMG1NbWorCwMGzxBIKonrFn\nZWWhubnZ+XNzczOysrK8jmlqalI8JlRoiTc5Odn51XHSpEngeR7t7e0hjdMX9HR9taDH62u321Fe\nXo67774bU6dO9Xpcb9dYLV49XmMASElJwS233IKqqiq3cb1dXy1EtbCPGjUKV65cQUNDA+x2O/bv\n34+SkhK3Y0pKSvDpp59CFEWcPXsWycnJyMzM1G28ra2tEPu3Hpw/fx6CICA1NTUc4WpCT9dXC3q7\nvqIo4r//+79RUFCABx54QPIYPV1jLfHq6Rq3t7ejs7MTAKuQOX78OAoKCtyO0dP11UrUb1A6cuQI\n3nrrLQiCgPvuuw8//OEPsWfPHgDA9OnTIYoitm/fjmPHjiE+Ph5z587FqFGjdBvvhx9+iD179sBo\nNCI+Ph6PP/44xowZE7Z4X331VZw+fRodHR1IT0/Hww8/DLvd7oxXb9dXLV69Xd/q6mqsWLEChYWF\nzkW9Rx991DmD1Ns11hKvnq7x119/jW3btkEQBIiiiDvuuAMPPfSQrjVCC1Ev7ARBELFGVKdiCIIg\nYhESdoIgiCiDhJ0gCCLKIGEnCIKIMkjYCYIgogwSdoIgiCiDhJ0gCCLKIGEnCIKIMv4/MOBWjpcI\nbLQAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "model.fit(X, y)\n", "y_pred = model.predict(X)\n", "show_predicitons(X, y, y_pred)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Tree-based models" ] }, { "cell_type": "code", "execution_count": 333, "metadata": {}, "outputs": [], "source": [ "from sklearn.tree import DecisionTreeRegressor" ] }, { "cell_type": "code", "execution_count": 368, "metadata": {}, "outputs": [], "source": [ "tree = DecisionTreeRegressor(max_depth=2)" ] }, { "cell_type": "code", "execution_count": 369, "metadata": {}, "outputs": [], "source": [ "X = newdata[['distance_log']]\n", "y = newdata['cartage_per_distance_log']" ] }, { "cell_type": "code", "execution_count": 370, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "DecisionTreeRegressor(criterion='mse', max_depth=2, max_features=None,\n", " max_leaf_nodes=None, min_impurity_split=1e-07,\n", " min_samples_leaf=1, min_samples_split=2,\n", " min_weight_fraction_leaf=0.0, presort=False, random_state=None,\n", " splitter='best')" ] }, "execution_count": 370, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tree.fit(X, y)" ] }, { "cell_type": "code", "execution_count": 371, "metadata": {}, "outputs": [], "source": [ "def show_predicitons(X, y, y_pred):\n", " plt.scatter(X, y, color='r')\n", " plt.scatter(X, y_pred, color='b')" ] }, { "cell_type": "code", "execution_count": 372, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAd8AAAFJCAYAAADaPycGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xt8FPW9+P/XzG5uuyGEXCCQBLyhEtBaW6loqXeUoKc9\nthZqS1v9qdXH6fm1R1uxytG0X7DFXv329KJy0Pa0Hm/taY8CKl6qiFat9UIIKhaFhCRIEkLY3dx2\nZ75/THbZJDOzs5vdzV7ez8eDh2ZnLzO7m7zn85n35/1WdF3XEUIIIUTaqJO9A0IIIUS+keArhBBC\npJkEXyGEECLNJPgKIYQQaSbBVwghhEgzCb5CCCFEmrnT9UIHDhxO10sl3bRpHg4eDEz2bqScHGdu\nyZfjhPw5VjnO7FJdPcVym4x8HXC7XZO9C2khx5lb8uU4IX+OVY4zd0jwFUIIIdJMgq8QQgiRZhJ8\nhRBCiDST4CuEEEKkmQRfIYQQIs0k+AohhBBpJsFXCCGESDMJvkIIIUSaZX/wDQRQ398NgeyvhiKE\nECI/ZG/wDQbxrl5FxeKFVCw6lYrFC/GuXgXB4GTvmRBCCGErbbWdk83bdAueu38V+dnVujfys3/N\nusnaLSGEECKm7Bz5BgIUbd5ouqlo8yaZghZCCJHRsjL4qvs7Ufe1mW9rb0Pd35nmPRJCCCGcy8rg\nq82oQautM982qw5tRk2a90gIIYRwLiuDLx4Pg0uXmW4aXNoIHk+ad0gIIYRwLmsTrvxNawHjGq/a\n3oY2q47BpY2R24UQQohMlbXBF7cb/5p1+G++DXV/pzHVLCNeIYQQWSB7g2+Yx4N29DGTvRdCCCGE\nY9l5zVcIIYTIYhJ8hRBCiDST4CuEEEKkmQRfIYQQIs0k+AohhBBpJsFXCCGESDMJvtlIehgLIURW\nk+CbTaSHsRBC5ITsL7KRR6SHsRBC5AYZ+WYL6WEshBA5Q4JvlpAexkIIkTsk+GYJ6WEshBC5Q4Jv\ntpAexkIIkTMk4SqLSA9jIYTIDRJ8s4n0MBZCiJwgwTcbSQ9jIYTIanLNVwghhEgzCb5CCCFEmknw\nzVaJ1neWutBCCDHpJPhmm0TrO0tdaCGEyBiScJVlLOs7Dwfpv/ZfLDOgpS60EEJkDhn5ZhOb+s4l\nv91AxekfNR/RSl1oIYTIKBJ8s4hdfWclFELR9ciI1tt0i6PHSV1oIYRIPwm+WcSuvvNY0SNa27rQ\nM2fBQL+MfoUQIo0k+GYTm/rOY40a0Xo8DF6wxPR+Su9BKs45k4rFC+Gb35QELCGESANJuMoyY+s7\ngzHlPFak01EwiLfpFoq2PIEO4HJBKITuLUX1+1B9PsBIwOLOO/H2D0kClhBCpJiMfLPNSH3nnq0v\n0/Pia/R/5QrTu4U7HYWznF2trSiMXBu2eXpJwBJCiNSTkW+2Gqnv7F9zB7gLzDsd2WQ5K36f6e3h\n6WqpHS2EEKkjwTfb2XQ6Ulv3WmY5W4lMVwshhEiZCU07v/nmm6xcuTJZ+yImItzpKKrAhl2Ws146\nxfT2wQsukDaFQgiRYgkH33vuuYfVq1czODiYzP0RyWSTHT2w4nIC11xHqG42OqC7XAAUbXlCyk4K\nIUSKJRx8Z8+ezc9//vNk7otIAX/TWiPI1s9Bd7kI1c8hcM11+L/3ffxr1jF4wYWRRCwAV2vruCId\nQgghkivh4HvhhRfidssl44w3Jju6Z+vLxlIit9tIyHrqCdOHJTXrWTopCSHEKGmLntOmeXC7Xel6\nuaSrrja/Rpo9psCcGaNv+seHYJGQ5Wpvozrog+oZptsdCQbhW9+CP/8Z9u6F2bPh05+GH/3ICP6T\nKPs/T2fy5Tghf45VjjM3pO0v4MGD2Tvqqa6ewoEDhyd7N5LPXUpFbZ1RYGOM0Kw6etylMIHj9q5e\nNaqTEh98AHfeSWCSC3nk7Oc5Rr4cJ+TPscpxZhe7EwgpspHPbBKywkU6EiadlIQQwtKERr51dXU8\n9NBDydoXEa27C3fLDoIN86GyKmUvE12u0tXeRii6SMcEOOmkJIU8hBD5SjKmMs3AAOWN5+He2QKh\nELhcBOc10LvpaSguTv7rRRXpqA76jKnmJKzzDa8xNpvSlkIeQoh8J9POGaa88TwKmrdHajAroRAF\nzdspbzwvtS/s8cCxx44OvBPJUk7llLYQQmQ5Gflmku4uY8Rrwr2zBbq7UjoFHRHuhLR5I+q+NrTa\nOgaXLjOmouPIUo5MaW98DLVjH9rMWgaXXTzhKW0hhMh2MvLNIO6WHcZUs5lQyNieBkc6Ie1F0TRc\nrXsnVnhDGfNfIYTIcxJ8M0iwYb7Rb9eMy2VsT7UkZilHgnhbqxHE26R6lhBCgATfzFJZRXBeg+mm\n4LyGtEw5O8lSdkSWGgkhhCUJvhmmd9PTDC84Cd3lijQ8GF5wkpHtnAZ2nZDiyVJOWhAXQogcJME3\n0xQX0/vMNrqad9H7h0eN/z6zLTXLjMzEk6Vskw1t286wxIOWjsQxIYTIUBJ8M1VlFcHFZ6Unu3kM\ny05I4SzlYBDv6lVULF5IxaJTqVi8cHwbQpsgrvoO4729CXXnDtSdLfFNQXd34d76nJH5LYQQWUrR\ndV1Pxwtlc53OXKkzGsu44wwEjEpUM2pGjXjH1WwO3/2a60bXbO7poXLBcagmvYGjv3R6aSkDK76I\n/3vft17KNDBA+ZKzcL/7DmjahIqP5O3nmcPy5VjlOLOL1HYWifF4jBKQY6aanSZSlTbdbBp4wVh1\nFP6n+nx41t9lnQU9MEDl3NkUvL0TRdPSW3xECCFSQIKviIvjRKpAgMKtz8f13EWbHjOdgi5fcjbq\n4IDpYyLFR4QQIotI8BVxcZoNre7vRO1oj+u51fb28VnQ3V2439lp/aA0Fh8RQohkkeAr4uMwG9ou\nSFvRZs0at5TJ3bID7NISVDU9xUeEECKJJPiKuMXMhgbbIG1lsPHicQ0Xgg3zQbWo+gUEjz9hUjLC\nhRBiIqSxgohfVBtCs2zosEhjhU0bUdv2GqUzQyF0rxcGh1CCwwDopVMYWHH5+IYLwSDen/4QXVVQ\ntPG7oRUV0/vkc/b7Ggig7nkfUNDmHCXdlIQQGUGCr0hcOBvaykiQZjiI5957Ik0jFL8fgP7PfZ7A\nv15vGRTDtaHH0gG9qIju7e9aLzMKBvHe+h2KH/g9is9nPG5kSRO/+o+4DhOwXHYlhBCJkGlnkVqB\nAEVPPWG6qfDll61HozZLmhRAHRyk/FLraW1v0y141t+F6vONW9LEt77lfP+dFBQRQog4SfAVKZVo\njWe7x4VZLjMKBCja+Kj1A//nfxxX1Up6e0UhhECCr0ixRBs1OMqWtlhmpO7vRG3fZ/24vXspven6\n2KNX6cwkhEgRCb4iteJp1ODwcREWPY61GTVos2ptH1rywP0xR68Jd2ayaTghhBAgwVekgf+qaxk8\n+1xCs2ZZL00ye9zIkibNXWC63bLHscfD4LJLYu5XrNFr3KP2sdeHTz+F0uuuhr6+mPsihMgvEnxF\n6vh8VJx4NFULP0LhX55BbW9HK51Cz+NPG1nQVk0UwkaypbubdzF8/PFx9Tj2N61leP4C26eP2Vc4\nzlH7uOvDnZ2U/OFBqk48Gu/N35YkLSFEhCw1EilT8bEFuA72jLrNdaiXik8upOedD+J4ogp6X/ib\nUWqyZYcx1RyrsMbQEGqMEafdNeewyFrlzZtQ29vQZtUxuLRx/KjdLjs7OIxn/V0ohw/jW/cTWaok\nhJDgK1KkrRV1TOANUw/2QFsr1NXH95zhHscOOMmWtr3mHOawoIiT1yt+8H4Kn/8LQ2edjW/NHVBW\nFvM4hBC5SaadRUoUPvfMhLZPlG22tMtF4Mqr7K85j02aMmuvOPb1YoyiFcDV0U7JA/dTecqJsl5Y\niDwmwVekRLBy+oS2T5hdtvTXvob/Bz8xv+acaFENj4fBi5zXslZ9PlkvLEQek+ArUkJ123+1Ym1P\nBqsGENx555E7dXfh3vpcpFjHRIpq+NeuY3jBSXHtY9Gmx1B37pBlSULkGUXX7fq1Jc+BA4fT8TIp\nUV09Jav336mkHmd3F1XzjkEx2aQDXTt3p68b0Zi6zNXVUzjQeoDyxvOMKlmhkLFm+IQTUfsO4Wob\nf+02VFdPzwuvxr5GHAzivflbFD/0IErAqGFt9h6E6QCKglZXz+DSZcZUeKwscIfy5XsL+XOscpzZ\npbp6iuU2GfmK1CjxoFsEEd3thpIUZfyaFbiIagAR3lbeeB4FzdtRQiGj9nMoREHLDlSTwAugtrU6\nq4rlduO/42d0t/yDnudeYuCzy23vrgCKrh8ZYa++UQp0CJEHJPiKlFD3d6JYBColGLRfX5sIq2u1\nfX2ou97Fu+qGyDZOOAH3jua4nl7BWVWsCI8Hbd58fD//lVEopNT6DDhayW/uNfb/k6dR+q/XSoEO\nIXKUTDs7kCtTILEkfdp5/nEo2vhGvLqq0rXjvaROO3tXrzJtP6iVlqKMdDZyQsd+mjhUP4eerS9b\nTz9btR7s66P0lhsp3LYVtb0dtJCjfdJKSxm4fGVC09H58r2F/DlWOc7sItPOIu3Uvj4wCbwAaFrM\nAhhxsSlwocYReAFwuei/+J+wOiO1rIo1duS96FSmfHkF6t9eMaaQy8rw/fzX9LzwKj3PbovdNCJq\n/yUrWojcI8FXpIRWVgYul/lGl8vYniROClw4pU0tp/CN1623W1TFGpcl3dFO8eObqGg8n8rj6vDe\ndINxvdjjQZvX4Kj2dDTpoiREbpHgK1JC7eszsojNhJI78nXUftDscYo6ql50qKISV083rrZWy9Gy\naVUsu9KSgBoM4tlwz6jR66hlUKqxH3Zi1qGOl3ReEmJSSfAVKaHNqEGrNy8fqdXXx6wGFRcn7QfN\nlBTT9fzL9P7hUbpefQu8XtO76UCotv5IJ6YxgcvpyLvo0f+F7i7jsUND+Neso2fry/S89Hf6v3yl\n7WOd1KF2JNEiIkKIpJLgK1LD42Fw6cWmmwaXLkusuYDNaG1sQQ0n2cVKIEDF5/6Joic2oQ4MWAdQ\n1UXv/Q/jb1qLt+mWcYFLq6xyNPJWO/ZRce4nR2czB4NoRx+Df+0626xoR3WoHZhIEREhRPJItrMD\nuZJ5F0vSjzMYxNt0i3lHoHgydyPPsxF1XxtabZ11QYpwtnFlFd6136XkN/9pmnE9VuCrV1L8yEOo\nPt+4beEMZ+93/x3PvfeMf+w11wGYZlvHMi6bua+P0tU3Urh1K2pne+LvGSafZyBAxeKFuFr3jrtv\nzCzuDCe/o7klV47TLts564NvIAD79yvMmKGn7O/GZHwRuruhpUWloUGjsnL0tkAA9uwxrkrOmWMc\nd6z3Iby9rEynr+/I/aJvLyiYgtt9OPJ4p88Z670/f2Efb31QxdyaHtY/OIU5c4yvXPgYpk8/sk8w\n+jkDAfjwWz+l5JEHOUwJD/IFGtjBpfwZzzXLOXDzunH3D/9cun832ifOYytncpgpdFPOP5jLRTzO\nR9jODhbgZogghXzM8zbTA3sIUMIO5tFGHXW0MYP97Gi8npnFXez+406GKKCSLrqpYoACihmivNLN\nnn//OVM3PsLpT63jgF5BC/PYzVH4KOVkmjmdv/Icn+KvLORY3mcZG+mgDjdDtDOTD0+/hKOuv5ii\nIujoUNj9ziA7/q7zxa8GObbBm9D3e+z3Vn1/NxWLTjVf/uVy0fPia5FiJNkmV/5YxyLHmV1yMvgG\ng9DUVMjmzW727VOprdVYujRIU9NQsqrzRaTzizAwAI2NJezc6QpXPWTevBCbNvXjdsOttxbywAMF\n+HxG4Cot1ZkzR+PQIYX29vHvQ/T71Nqq4nIZeVB1dRrl5ToHDyrs2xe+XaG+PsSFFxrX/554wvy9\ndfre//i7naz7xbGMXzmrjzyPcbuigK4bxwIQCCjU1mqUlel88IGK32+e/jSNbryzymnvdFFbqzF1\nqk5vr/E+zJql4fNBb69i8vpmNCrp4hDlBCmM4xNLjxkzND796fi+3zLyzT1ynNklJ4Pv6tWF3H13\n0bjbr7lmkDVrhpL6Wun8Ipx7bgnNzeP/ui5YEOSMM0Kmx2wm/D5YvU+JiPWcY9/76dO9SFpBcsXz\n/Tb73loVIwlccx3+NeuSso+TIVf+WMcix5ldcq7IRiAAmzebn/5v3uzO2tUT3d2wc6f5kpOdO138\n7/86H9Jv3uymu9v6fUpErOeMfu/PP92HsxGniMdEv99WnZ5sexsLIZIuyRO06bF/vzFVaqa9XWX/\nfoWjj07LgD6pWlpU66WxIejsdH6u1N6u0tKiWr5PiYj1nNHv/Vu7K5L2uuKICX2/AwHUPe/T/8Wv\n4P+3b6P29Y0vgymESIusHPnOmKFTW2uewTprlhZJ3Mk2DQ2aXVEoampiZ+2GzZql0dCgWb5PiYj1\nnNHv/cnH9CTtdcURCX2/g0G8N3+bygXHUXHWIirOOp3K006m5J5fQWHmXd8WIh9kZfD1eGDpUvOi\nAEuXBrP2RL6y0kiuMjNvXoh/+ifnhRCWLg1SWWn9PiUi1nNGv/dPPaOCZYVkkahEvt/eplvwrL8r\nUudaYaRm9Pq7ZH2vEJPE1dTU1JSOFwoEkpsE9alPhTh8GD78UMHvV6ir01ixYpimpiHUJJ9SeL1F\nSd9/K8uXB9myxUV3t4KuGyPe+fONbOdzzw1x6BC8+67K0MjulJbqzJ0boqhIN30fot+nvj4Fl8vI\nLK6r05g9W6OgQOfwYWVkxG1kO3/+88OcemqIAwfM31sn773a1krp+pt4lk+bHOUQbrca6bugKEeO\npbBQj2Rj19drBAIwPGz+Xk2dqlFdfeS4Z8/WKCw0fq6t1VBVnYEBp9edNSroYdhdbNkPYvLozOQD\nLl8epOkHiuPvt9dbRKCrlyk3fQv1sHk5T3X/fvpXXgEFBUnc3/RL5+/oZJLjzC5er3Wya9ZmO4fJ\nOt8MXecbCFC54DhUn4+P8Th/52zm8QoPcS1Hebvo/ttb7PnQKOcYa51v+Fh9Pp0//amAE04IsWyZ\n8b6M3YexP3d3w1+fG0a76bv09oYs1/mexqtUug7R9szf2TlwDO3tCrNm6VRX67z/vkpN9RB7/88D\nhF57i+qedzkw7Xj8J5+GvvwyKqoUejqHKOnv5aRPevH9cAO7/7Rj3DrfZ90X8HLwFI4u7eaCJSE+\n+Odv4i5S6OxUOHRI4YQ5Aaq/dQ2dH7p4l2N4k1O5irs4nj1Mry1gYNtzcV2fra6eQvcrb1Jx+kdR\nLH7NddVFz0vZu743LFeyY2OR48wuObnUKJ1y5YsQS1KPMyr4jqWVTqG7eVfaEn3siktEs1vrarlE\n56tXogwMULj1edSOdqP61oVLASh64nHU9jb0Eg+qb/z7ara8J5lLgaqrp3Bgz34qzvw4LovSmaG6\nenpeeDXrk67kdzS35Mpx5txSI5H51P2dKH6/6TYl4E9uh54YnHY9sqyfbNO1qOS+DZQ8cD+ufW1H\naiWvvwtU1Wia8Mw29KlTTR9r1iZw3FKgunr6l1+O/8YEr816PLbtCwcbLzbt0iQdj4RILQm+IiW0\nGTVos2rNt82aNbpDj90f+2QEApuuRzrEXOtq17XI6opy0eZNxv8UF6O27zN/3n2t409C3G6j29Gz\n2xi4bDmEQhQ/9N9UnHV6wt2H/E1rCVx5NZq7wGifCOiKwvD8BfhXf/fIHcd2PDrz45T+/9dCEts/\nCiEMEnxFang8KF1dppuUA13GaMuuvV2SW99FjygJF5e44mp6tv2Nnq0vG1O6FnUbtcoq9JL4pmXD\n/Xe1yirLXr26x2vZJtDbdLMxou5oR9F1XPvajO5D3/l2XPsBGMfldqMGhyPZzoquU7CjGe+a26Je\nc0zHo31tlDxwP5WnnChtB4VIsqwssiGyQFsryuCA6SZlcADaWvH++j9GXd8Mt7cLs9qWUBnEkRGl\n/+bbqA766BlWHReZ8N6xFtU//tq1Ha3GGN17v/vvqPEErYEByi86F3dLs+nmkt/8J7gU/GvuMD9Z\nCHd1KiuDvg/BXQpgOW1etHkT/ptvs72P6vNN7L0XQowjI1+REoUvbbPf/twz1gFh42MUbXrMfJvJ\nddL4dqwQfv5zKpac7WxEbXO9187Q4sUj+/uo5X0Uv2/0tHMgQPn5n6SgpdlyOlsBPBvW4731O6M3\nhGcKPnkaFZ84haoFc+G446hYfBqlq65HbWs1fb7wCN1uaj3M8Xsv14yFiEmCr0iJoUVn2m4PHnOc\n5R97tWOf9XXSkWCRKG/TLXDnnY6byTsJSmNppVPwffd2Sm/4Bmqn9b7qRcXGCDUcOM/4GO5333X0\nGsX3bRh1LTYyZdzWakwrj9QpdbW2UvLg/eher/m+zqozrs87SEob9d53d+He+hx0R11aSPKlAiFy\nmQRfkRp19WjTpplu0qZNQ/vIRy3/2GvVM9BqZppvGwkWEfGMsmxGsVajOqeZ0tFCc+ZQfunFlPzh\nQdvWEspAPxUXnEX5krOMwNm+z3ErCjU4TOlN1xs/JDg6h6gMb5uktDBtVh1aeTnl555J1YK5lH/2\nEuO/554JAwPjrxnHOLERIp9J8BUpo9XMsr7d5o+9ur8Dpe+Q6bZIsEhglGU3irUcUTsISmMV7Gim\nYIf5NdtoCuBqa6WgeXtczx9W+MLWSLMEq2nlyGv5fEaWs8tlZHjXzR6X4R1OStNKzdcmDi5tpPyz\nl1DQvB0lFIqMsAuat1N+0blxn9gIkc8k+IrU6O7C/e7bppvc774N3V2j17RGbQ/XHgZjCtes9V0i\noyy7Uey4EXUUq/2cbOqBD5nyL1dT/oXPGTVDbUSynEeC5uAFF47P8B5JSut+6TX6l/0Todq60e/9\nv/4b7p0tps/vfrsFtXWv+X5O8FKBELlIgq9ICXfLDuz6I7pbdhxZ0/rks2gzzUfJenk5Pc+8MHo5\nUALTx4DtKHZUgY2xU9nR+2kRoFMhvCbXkqZRvPHRuKarw4qeenL8+xSeTWg8n+LNRsLbwGXL6Xl2\nG/4163C/+471Z6ppaNXV5ptsTmyEyFcSfEVKBBvmY9kBQFWN7eEf+/osR0ZqRzsUl4xaDpTQ9PEI\nf9Na+MY3zJvJx5jKVvv6UA98GOvQk0tV0YrMi7Nb1mvGqNkc/n/TpzV5n6zW+XrvMGYbgg3zset5\n6ejERggBJBh8NU3j1ltvZfny5axcuZI9e/Yke79EtqusQiszL6uolU2FyqojP8c5HZzo9DFgjJx/\n9jOj9OOLr40aUceayk4k+copnfEjXQVQNA11cHDU9phT36qLns1PwyuvoNWYvxdazSwYGDgy+rWb\nTdj0mHG/yiqCJ84zvU/wxHn4f/CT0aUxY1QOEyKfJRR8n3rqKYaGhnjwwQe54YYb+MEPfpDs/RLZ\nLhBA7T1oukntPTh6ytPpdHCi9zfj8RidfKKmmmNOZXs8DF6wJOZT6yj0f36FEYRUlVD19JgBU4n6\nF2t7rClmbcZ0Sn5/H3z+85ZLnZRDB6k454zI6F7d12Y9m9DWamRWB4MMf2KR6X2GP7HoyPS8yYmN\nEGK0hH4rXnvtNRaPFBE45ZRTaG6Ondkp8ov6yksxt2tnnxf5OTw6Ktq8CbW9DW1WHYNLGy1HTfHe\nP+b+xprK3vMBFBfT/6WvUrJhvW0A1Opn47vjZ5Hn1crKqFhyNi6LhKRkUzs68Pz2PmB0oNYB3e1G\nDQYjCW2RymHDQbTaOtN9VICSB+5H93go2vKk6WsWbXkS/62ByLKlbG9RKESqJRR8fT4fpaWlkZ9d\nLhfBYBC3zRnutGke3G6L60VZwK41VC5J2nG+t9NykwJUvrcTLvvM6A13/dIYYXZ04Jo5E4/Hg+0Y\nNt77Rxl3nN65MHs2fPDB+P31eqlc+Xloa4O6OpgyBQ5btztzXfoZqufMMH4I//fSf4Y773S4dxNj\nO3q2WIrleXYLLGuEX//a8nk9T2yGjg7Tba72NqqDPqieEefeJo/8juaWXD/OhIJvaWkp/qh2cZqm\n2QZegIMHs3edX670lowlqcd51hKquNk0EOhA11lLwOq1yqaDPwR+h/sS5/2tjtO7ZKlpL136+o5U\nk9prPXrVSqcwcPmX8K+6bfyxrboNb/8QRY/+GbWjPe7s5FTT9+5l8INWiu3u096OVjMTV0f7uG2h\nWXX0uEutP9No4frTDupqOyW/o7klV44z6f18Tz31VJ5//nkA3njjDY4//vjE9kzkrrnHoxcWmm7S\nCwthbuZ9Z8x66WpRMzzRtNIphGrrjfvV1tO/4nK639hpfY0zfD30mRfQZppX75pUuk7R4zGqZCkK\nmkWZSkfX2s2yyW+6HvW9XVKEQ+SdhEa+F1xwAdu2bWPFihXous7tt9+e7P0SOWDgshV4fv9b09sz\nUlTnI3V/Jwz0U3GOeY1qpT/AwY1boLg4vhFcZRWDl3zGfIQ9iayWLY26j6ZR8N6uUcljusfD8Mc+\nTv8556HubEGbc5TlexHOJg9zte7Fs2E9JRvWo9XWMXjhRfRfdZ3RB1qWJokcp+i6g9+6JMjmKYRc\nmQKJJanHGQhQdVw9SnB43Ca9oICuXa2T9gfW8XEGAlQsXmiahBSqn0PP1pdjH4PZFGswiLfpFoo2\nbURt24txNdb4Ncy06WgnRgXj4mIGvvQV/N/7/ugZAJv3cuxzafWzGVy6zEieGymqEmuaWn5Hc0uu\nHKfdtLOsARApob6zE0wCLwDDw6jv7ET76MfSu1PxGlnSZDZKjTnNGg6wmzei7mszRnZRAcW/Zh0M\nB/Hcew8OVu5mtOgTBmVgAM/6uwDw3/7DyO1Ou0MpRGVgaxqoquV7KEQ2k2+wSAk1utWcxXYtTfsy\nEYkuaTKdYo1uSB8IUPTUE6nb8UlW/MDv8a/+buQERSsrQ5sxA5dFtrT5c9yP6jsy+hn3HgqRxaS8\npEiJYIxRbaztGSORwhEOCnYk0ic4myg+H+o/3juSZLXkbNQ4Aq/xHObTjtIlSeQCGfmK1KisIjh/\ngWlrveD8BaPKS2aFOApHOKk9HS5Vma7CG+mmAJ5f/hy9ssJ02l4n8evbkfdQCnmILCYjX5EyvZuf\nYXjBSeg2oOWaAAAgAElEQVSqOlLsX2V4wUn0bn5msnctpRzVnk6gT3C2KXxxq1EX2oTu4JqtVXDW\nZtSMrkstRBaSka9IneJiep/ZBrvepXjL4wxccFFGru9NhaEzPknJg/ePuz06UWv89eRatKlTUXsP\noXbsQ6uejtrZkZUZ0DDSkcpqWzBomWaml5ZGyl+aPra9nYqzTkerr2dw6cXSuEFkJVlq5ECupL3H\nkvTjHBigvPE8owF7KAQuF8F5DfRuehqK7WoppVbKPs/oDOe2VnSvF1BQ+gOjE7XGjvrGLqUZ+ZmB\nASrOOj1rg2+iU8taiQe13/moNnDF1Xg23C2/ozkkV45TlhqJSVHeeB4FzduP3BAKUdC8nfLG84wR\ncY4Zm+GsjIze+ldcju8HP7FemjT2enL450DAuC4cIzFrItdPUynRfVLiCLwAJb/dACUFsHqNLEES\nWUOu+YrU6O4yRrwm3DtbIMZSpKxjk+FcuC3BE43CQrRp02LfT7X/Nc7uVcSxKaEQ/PKXkb7LEYEA\n6vu7obvL+K9cIxYZRIKvSAl3yw5jqtlMKGRszyFOMpzj5W26ZfTMgRUtG1ZMO5foiDmyBCm8vOmT\np1HxiVOoWjCXik+cQsXi0/CuXgUWnZ2ESCcJviIlgg3zwWXRQtLlMrbnEEcZzvGwGUmH6UD/55aj\n1dXb3i/dU9KTNdIOn+SEp/9dba1GG8VQaKRyViueu381foQsxCSQ4CtSo7KK4LwG003BeQ3Zt843\nFpulQ5EM5+4u3FufczTl7qQIh1Zbj+9HdzLYeHFCuzxWsoKmk2CfigCtzZyFVlYW86SlaONjMgUt\nJp0EX5EyvZueNtb5ulzGOl+Xy1jnu+npyd61lBjXkrB+DoFrrsN/079Tfu6ZVC2YS/lnLzH+e+6Z\nxlpVC3Yj6bDBZReDx4O/aS39n10+4YA20RGyDmge85aDyX4tM0NnLkbt60Nta7W9n7qvldJrrzzS\no1mISSCpgSJ1wut8u7twt+wwpppzbcQbbUxLwvDSofJzz4w/69umqYPmLmDgq1eOXt+qZEZalRLw\nT84LT5mC77u34123FhQFbFZQKkDJ45soOmEOA1/6Kv7v/1CypEXaychXpF5lFcHFZ+V24I0WXioU\nnmpOMOvb37SW4QUnjbtdDQ4bGc4jAcPbdAsljzyUEcuNkjF6TsiVV+L9yTo8996D4jABTQ2F8Pzm\nPym/4FOShCXSToKvECk0oazvoSHUQ72mm4o2bzSuWzpIzEqXVAd/feSfVlqKVjoFXVUjU/t873sJ\nvw8FO5rx3nS9LEcSaSVzLUKkUCTr2ywAx8j6Vvd3oraaX79UW1sjy5dyqTuSXQAPb4sUL/n8Cnx3\n/AwAz9/+NqH3oeR3v6Xkd781egaffyH9V33NuOZu17M5F4ytribSRoKvSL18/gUfyfo2W68bK+tb\nKyujS6lmk34Bz/ApWpnNPFqYxX6q6KbjkRp2t3o4wdNEga+Hcg4xRCGvs4B26qhjHx9hO0MU0Mx8\n3Ayxh6MZoACVEAeoZgp+Ps4rdFBPGQfpooZBXOxiLkUMMAUfpRziQ+r4OC/RQw0+CmmjHh8eyuil\niCBFDFPJATqppZB+igjio5g+plHJfnZxIn14KWSYKfgooZ9eKtAJMUwhHob5FM8wRCGvchr9FKOh\noANF6HjwM4c9rOE2zuFFAIo2PYZeWkrRliehrdWy2IiTCmDhqWpX6148995Dyb33oNXPZnDpMvOS\noNkuuhTqvjbjpCNXjzVDSW1nB3KlzmgsST/ODP0FT/vnaVfjWtNMT0wGBmDJ2S7e3l1MZhaPnEw6\nrdRQywFH78zw3LmoHx7AZTGFH0vgmuuMHs4ZIFnfXe/qVabJfJlyrLnyN9eutrNc8xUpEyl20LoX\nRdOMUUU+FjkYyfruat5F7x8eNf775HN419xGxeKFVCw6lYrFC0dVX2psLOHt3R6MX1FF/o36p1KP\n84phBbt2oQwNOb7/WEWP/jm3yqHa5AlEqoSJlJPgK1JDfsHHi8r6tjsx6e6G5maL6mBihMJfOMP5\nveNs1hBN7Win4twzc6Y0ZSpKoYr4SfAVKSG/4DZinJi0vD6c5h3KTndzreP7TmTiXgFcHR05M2uT\n9FKoIiESfEVKyC+4tVgnJgsq96V5j7LTNfw67a+ZE7M2TkqhipST4CtSQ37BLcU6MZl2wnQWLLBY\nGyxG6JGs5+Q/s3Wxj1yZtbEshRpdNU2klOSUi5QJ/yIXbd6E2t6GNquOwaWN8gtuUzoyfGKyaVM/\nS5aU8Pbbcu13PCPbOVW0mpkAuDo7xm/LlVkbi1KoIn0k+IrUkV9wS7FOTIqL4fnn++nuhi1bVLZt\nU2lrUznuuBAzZ0JVlc7hw7B7t4v6+hAul8LUKSHcP/4pb+2vyIt1vqmi+A4TOupo0+A7dMaZKX3t\ntAuXQhVpJ+t8HciVNWexyHFOgmQWIAkEqFi8EFfr3qTsmpPiFLnA6jiHF5yEeqgPtb0NvcQD6Ch+\nv7FevfHiSVmvnlHf3RTKleOUdb5CZKroJgwT5KQHsBjP6gTD9cEH9PzhUQYvvAjVdxjV50PRdVxt\nrUbm883fMu4YCEhdaBE3mXYWIkeEE7mSNfLNh1GvHdV3mMqzTrdcI1xy3wYK/voSqt+Hum8f2owa\nBi9ahn/tOinRKGKSka8QucImwzydMqOzcHKo/QHLkxAFKHh7J67WVqNQSkc7nnvvoXzJWdbFOGSU\nLEZI8BUih0QvIUFV0UqnGO33XC5CtfVopaWmj9M8XkI1M5MTOC0aHOSLgubteFffOPrGYBDv6lWW\n5URF/snv3xIhcs1IhnnP1pfh3Xfpbt5Fd/Muel58jZ5trzJw+UrThw186cv0PLsNrSYJy2gcNrPP\nZWOLcUidczGWBF8hcpHHA8cea/w3KqnLrriC2teH+uGHE37pfL9WDKB2dhwpxhFPnXOZls4bkhUg\nRD6xWXud7IStfKbVzIwU47AtJ7qv1fgc6mdnZPtNkTryqQqRj8yKK3g8DF6wBM+G9Ul9qfB15Hwa\nEUeXULU9qVEUSn79C3CpeNbfFbk5PC0NZER/XZF8Mu0s8oNM59kbSQgq2vKEUdvY5bKtcRyPcBfe\nfKG5C/Df3HTkBpssdCUUwnPvPRQ/8HvT7TnRyEGYkuArcptJlinf/KZkmY5xJCGo1QiWoVDMgKmj\nEKqoTMfuZRVFC6F2d426zd+0lsCVV6G7zGt1Kz6f6e250shBjCfBV+Q0syxT7rxTskyj2SQEWS0b\n0goL6Xr+r/Q07yJw5dVoivwpCdM9HrTKqtE3ut30f+3rEGc135xp5CDGkd8YkbviyTLNY7ZlKS2W\nDSlDQ1R88TK8TbfgX7OO7l176f/MpYSqqpM2XZ2pYh2b6vNRsfAjMDAw6na7VpJWswz53n4zl0nw\nFTkrVtN6mc4z2PYXrptN4IqrjaVJUSFCgdFrVcvK8N19Hz1/287A8stz+hqvk2Nz9XRTftE5R3IN\nurtQ93cyeP6Fjl5Dd7kIXHmVtN/MYRJ8Rc6K1bRepvNG2CQEDTYuw7/ux/Q8+SzaTPP3a+wsQuGL\nL6RkN7ONu2UHFWd8jIpPnELVgrlUfOIUip7czPCCkwjVz0ZXVdtRdP/Xvi7LjHKYBF+Ru+yCikzn\njWJXfAMwCnDs32/62OhZBOmsdIQCuNr3jUpgc+1ro6B5O4MXLDEqitXVmz5WTg5zn5xWiZxm1rTe\ndeln8K+6bZL3LMPYFN8A+7Wq0YEiVYU63mA+a7mRN/kIXUwjRCEl+PHQz3TaOUAtXUyjjEP0Uk6A\nUjz4qORDPAQ5yBT6KEdBo5hBCuinEIV+CjhIFYX4AQU3IYYoopAhyumlnB72U08ZXXRTw0FKKSbA\nVHzU0s4hphIChvFwBi/yPdZwAv+IeTxFW7bgv3UNg40XR9bzRot5ctjdhbtlB8GG+TA2uUtkBUXX\n40y/S1A2N0bOlcbOseT0cUY1ra+eMyN3jzNKsj9P7+pVpoEicM11owpBWN0vEV1MZTofolOQlOdL\nB4UAPdRQjvnyITCu6fZsegr10CGKHv0TRX95NnJyOLi00bKyVfWUAoZPW4h7ZwuEQuByEZzXQO+m\np6G4OJWHlVa58reounqK5TYZ+Yr8YFbRyYmooJ3v09RmswiRQGF2v02Poba1Tij5qoYP0SmcwDOk\nn46XmXTSj3kHKQA9FKLiwnOMH1wugiecSO+zL6LNnmP/PVu0iILm7Ud+DoUoaN5OeeN59D6zLUlH\nINJBRr4O5MpZWCxynFGCwayvtZuyz9PhCYm6cwcVZ5+B4vBPjM7oTOI3mM9H2U521sfSeZu5jqag\nw4YXnGQfQLu7qF4w1xjxjn01l4uu5l3mU9BZeAKZK3+L7Ea+knAlhAlpAWcjqkuSHW3O0ZYJRWHR\nYXlsiH2QFYntX4Z4jEviur+7ZQfq669Zrj93t+wwDbwAhELG9mjSQzijSfAVYiwpzpEcNtnmTizn\ngSTuTPpdzKPxPUDTqLjoXPMgGQxS9Kc/WD/W5TKSr6LICWRmk+ArxBhSnCN5wkuYtFLr6Tcrp7AD\nF8Mp2KvUKyYQ15QzjDSg0PUjQfKWVZFt3qZb8PzXfZaPDc5rGD3lLCeQGU+CrxBjSHGOJBpZwtT9\nxk76V1xOqLb+yDriK66OOS3dyXQUhiBStDLz/yn46WDi35GS327Ae9P10NdnGUh1YLhhvpHtHBYI\n4H7tVdS2VtPHyAlkZsiOzBEh0mlkujSh9ZfZKB0JOWVl+P7vr8e/VoHbdllSFYfQKMq5db5OKKEQ\nng3rUXw+VKt10y4Xfff+3lhmFAzivWUVRY9vNIKrqppeI07aCWQWJnJlEgm+Qphwuqwmq01GRveY\nJV9m77PmKaHgnbdHPewUdvAwX0nNPmW44ocftMz31mbWGsEvGKR8yVmjlyFZmPAJZA6sBMgEstTI\ngVxJe49FjtNEFp/dxzpOp0Uz0iL8PpeVUbHk7KRXyMpVww0L6P3Li3hvuh7PhvWm99FdLtB1tNp6\n2wIeTqXje5Mrf4tkqZEQiXK4rCbrZFpCzsj7rPb1WSa75XKbwkSpB3ugu4uiTRb9mEf0Pvxnera+\nbATHiYxOM+17k8Uk+AqRhzI1oztWstvwMcemeY8ym7q/E/frf0ft7LC8jzZ9BsGPnRb/CWS4HWJU\nQM3U7002kuArRB7K2Ixuu05UF19C3389KCPgaJpG2fVfB8W6Ctjg+UviC7w2xTky9nuThST4CpGP\nMrjdol17Q622Dq1+9qTtW6ZRAFdnp2UJTx3ov/brcT2nbXEOm+/N0Blnxrn3+U2CrxB5KlYP30kz\nsja4Z+vL9Lz42pFrlUNDqPs7GbxgyeTu3yTRAaZPj2vkr5eWxjcadXBNd+z3RiudglZaSvFD/y0l\nLOMwoeC7ZcsWbrjhhmTtixAinayCXKYsFwknuxUWjpoGLdryBMMLTiJUN9s4aaibDaecglZq3UUo\nZzzyiO0U81iqz4f3DucnU46u6UZ9bwYuW47qO4zq842uziUlLGNKOPiuWbOGH//4x2ialsz9EUKk\nW4ZndI+fBm2loHk7gxdcaJw0vPAKvP463W+8Tf/nPk9oRk2k3lTO8fnQambG9ZB4spDtrunqRcVo\nHu+o2wq3vTDh18xXCQffU089laampiTuihDCLMM0r9lNgz715JH118Eg3jvWUvjyX1H3dxp1ktO7\np+nR2IjSezCuh8SVhWxzTVcN+Klc9NHItLJkPk9MzCIbDz/8ML/5zW9G3Xb77bdz8skn8/LLL/PA\nAw/w05/+NOYLBYMh3G7XxPZWiFwVDMK3vgV//jPs3QuzZ8OnPw0/+lHmTANPhn/8A44/Hsxm2Fwu\neOcdOPZY+Jd/gV/+Mv37lw3KyqCjw/nMRl8f1NXBYZsiF9/4Btx+O8yfDx98MH77UUfBjh0ZO5uS\nCWL+Vl922WVcdtllE36hgwez90w+V6qtxCLHOXnGVQ364AO4804Ch/30f+3rCVXYysTjjJu7lIra\nOtOKV6FZdfRQjPfKa/D8dsMk7Fx20DSd7r37Ufv6HH2P1Pffp8Lvt505CP3xT/T823fwLllqXu1q\nyVL8/hD4E/v+5cR3F6lwJURms5laLfnNvfndCD3GkijvHWvx3HuPdZP5PGF3jVvxHabinDON79Gi\nU/GuusH2e2R33TcsPK2csRnzWSCP57OEyAx2186UkaASziIF0l93eZJZNrm48RYqzjljkvcuc+hF\nRSiDg+NuN9YCGxWwXB3teO69h4JX/0rvk8+ZX9Kw6eoVFimoMZL57L/5tqytgT5ZJjTy/cQnPuHo\neq8QwpqTkUZYXmaRWiyJUru7LE9awnQwRmNf+f/o//yKhJcjZXrmtAKoJoHXSkHzdryrbxx9YzjZ\nr68PNA2ttNTyuMcVYsnwjPlMJCNfISabg5FGWHi6L7otX94Y044wfNJidj1Yd7nov/zL9F/7L8aJ\nzUhQ8B3soWLhKbgO9Tp6Sa10CnphIa6e7vHbFAU1PU3h4qJ5vSgxrtmCcSLnv3WNsY46qkWg7vGg\n+nzj7q+joNXPzr3WmpNEgq8QiUhyq8FRU6v7WkFRIlPO0Uzr52Zx28MJsTlp6f/ylfjX/Xjc7d4f\nrzMNvKGp5SihIMpI0NFLSxlcejG+b3+HqkWnmr68VUnHSRcIOFpmpe7vRN3fScl/3jXqPVRMAi+A\nVjODniefhcoqy9fNy+9hgiThSoh42BSdn5DoqdWX/k7/l680vduo6b5U7UtYFqw5Dif8cNRRoxN+\n1ppcF7dJbKNsKt2vvkXPc3+l57mX6G5+D98v7sbdujfrkrmcjsa12nq0sjLr92Ts83Z2Utq0evz3\nK9XfwxwlI18h4hCuthSW9ESokalV/9p1UOAen2QUNd2Xsn0JBkdNQ2q1dQwuXTbhJuwpMXLS4vnp\nj+hp3mU76opZFKKvD21eg3HDyIlH8OhjjPXEWRaAnRhc2mjbP3ksBSh58H70qVNHfb9S/juRozLs\nN0mIDBaj6Lz/5tuSN90WK4vUyb5gvcbQTlb+MR1zPdiM3TXiyHS+yYmHNrXc9JqvXlSMMjgAwP9w\nEf/OGrqYhp9SVDQ89DGMl2J6UHHRRRXDlFBOOwNUUEYXhehoDNPBMbjx4cKNl4P0UQ4oTOMgpfjY\nTw3T6KWBN3iL0yihjwG8eAhQSIhpdKGjchzvcYBK9lODikYBOkfzHn1MYw7vU0snc6b5ePOoi9nV\nsZhp32nFq93BsbyLDy+v8xFO4Q2m0E8zDQTwchzv0s4s+ihjLv/g2Ae7aRnu5riPl+L1qJz76Dbe\nZT53czXl9HICuziG9+l5xM9fCnXKq1yEQgpTp+oUFsLQEPT3K5xwgkZREfT0KFRUGKP1nh4Fl0tn\naAhOPRVKSqClRaWhQaOy0vxz7e6G119X8Xp1ysthzhzd8tcwEIA9e4xJ+enTdfr6FGbMsL5/KsWs\ncJUs2bxgOlcWfMcix2lPfX83FYtORTGptqS7XPS8+FraEqGc7Evlwo/Ef5yBABWLF5oXtaifQ8/W\nlzPyep7Tz3RcMZMRgWuuw79mneX2UEUl6qFeYwTschE84UTUQ328v09jLnvJ0WKWk0ghOsfc5YJ5\n80Js2tRPcbFx28AALF1awo4doysner06X/jCMN/73lBkoiYYhFtvLeSBBwrw+YzPSlFA16GuTqOx\nMUhT01DSJ3bsimzIyFcIhxyNnLJ8X5zU683mTGvLNcNNa+2vCXtL6Xrqedzv7ybYMB+1r4+KRacy\nl2EkdSZVjpzQhELQ3OymsbGEZ57pB6CxsYQdO8aHML9fYf36IlQV1qwZAqCpqZD164tG3S887Gxr\nc3H33UYAD98/HeRbI4RTmdSAPtF96e7CvfU56O4y3Wy35jjdJxgpYdNGMeaJx/AwwcVnQWUV2owa\n/lj+RWTEm147d7ro7jammlta7HsFbNzoJhAwppo3bow9zty82Z3W3EIZ+QoRB9uRUybvy8AA5Y3n\n4d7ZYgwjVJXgifPoffxZIvN4YLt8J+0nGKlkco04rtkEj4f/CFyb6r0UY4RCxjVgMO+1Ea2jQ2X/\nfuPkqL099jizvd24/9FHp2cJmQRfIeKRSeX04tiX8sbzKGjefuQGTaOgZQcVp8yjp3nXqCzmTDrB\nSKt4TjwCAb5e8mv+MrAojTsoXC5oaDCirqraB+CZMzVmzDAC6axZGvv22Y+UZ806cv90kGlnIRKR\nSeX0Yu1Ld5cx4jXh6uk2Cu1Hs5mazXVOGwWo+zu59NDvyfzCk7ll3rwQlZVQWQkNDfbLv5YtC+Lx\nGL8Wy5bFXnO8dGkwrb/Oku3sgGQB55Z8O0731uco/+wlllcnQ9XT6Xn1rcw4kUhQ0j/TWNWaRrLC\n328NSbZzSumAItnOQojsE2yYbztHp3Z3HcliDgRQ97wPKGhzjsrqgDwhsdYNj0xRH3f3r9BxjVrn\n6xtZ5+vNsHW+R42s8z2a93mL+ezhaKbRjQudo3nf8Trf43mX95jLfHbgZYDTeJU2Zo5b59tLGX/l\nE1TQjYabqRyiiCEGKaQfDw20UMwgB6iiGiMBsIsq3AzxITVcwBY89PMmp1D7/AYq51aM+giKi+HZ\nZ/sdrfN1u+H224dYvXpI1vlmk3wbKeW6fDzO8rMXUdCyw/R+ofo59Dy7De8P/g/FD/x+VH3jgRVf\nxP+972f8lPOkfKaRghzmy5bKl55Lwa5307tPOUgHev/wqJFpbicDa0vLyFeIPNf7+LNUnDLPtFJT\npCn9+rtG3a74fJHb/Lf/MC37mVXsEt7cbtSBgcndvxTSAW36dNQPP4x7wt2YSI6D6jJmb6xkUznU\nKJJwJUQ+KC6mp3kXgZVXEKqejq6qR5KJbryFoo2PWj/0gd+j7tyR0Q0WJpVJwpvdmuFcoACD5y9B\nT0N/5GBDg3UnJY6UQ3W17kXRtEg5VG/TLQntW7pI8BUiX7jd+H98Jz2vvkXPS3933JRe8fmoOPsM\n6VYTB7tiJbmSH13y4H8Tqp8d9+OCDQvQHDxOB4bnNdD3y/XWJ34xapxn8gmjBF8h8s2YkZo2owat\nxrpylYLRuzZbRhQZwaYCWf8Xv8LAhY1oJSVp3qnkUkIhCna2MLzgJEK19eiKgq6ar6XVR/4Nz19A\n7+PPWL43EcXF9C+/HNXno+LsRZYnfk7KoWYqCb5C5DuPh8HGix3fPdNHFJnCcs3wD3/K4f96gO7t\nuxhYdknaRsKpeh31UB89Tz1Hz19fp/8r5n2oBy64kK6du+l99kUoLsbftJb+5Zdb79PAAJ4H7485\nlZzN5VAl+Aoh8K+5g+H5CxzdN9NHFBkjVrGSsjIO/+IetLr6yd1PC06DdaQX8owa+q/6GoErriZU\nPxsdBd3lQgcKd7bg/ekPj4xc3W58635iGThxmY+gx534xVvjfKRPcyacPErwFUKA203vlueNP5w1\nNSNTiOZ/HjJ9RJFx7CqQFRailZenZTfizUp2en9tZi0lv/4FFYsXUrF4IUVPPYE2ZQoKOkoohAK4\n2lrHj1w9HgaXXWL+pCHz6lVmJ36OqpIFg3hXrzL2cdGpGZG/IOt8HcjHdaG5TI4zhpH1kiW//gWe\ne+8Zv3mk920mydbP1Kp/sK6qpr2aM9HwgpNG1w23Ma4ntNla6QsuwPP0Ftizx+Txs+nZ+opl1TGr\ndb6x+jinit06Xxn5CiFGGxmp+deuc1Tn2FQGTe9lLJtMXW3mLPo//c+kK/zqQGhGjeOpZh0I1c0m\ncOVVqId6Hb/OuJGr2dT8D34C06aZPl6bOtW6gIbVDEOGZkRL8BVCmEukwcLY6b1FpxqNG2R50jjq\n/k7UtlbzbR3tBG6+je732ui/+NNohUWm97OjR/2LvTMueu++1/Jaa7ThE+cZ34cXXqH/a19H3bfP\n8T5ZXrKIDpyBAPT0mO9m76HxwTLGiV6mZkRL8BVC2Iujg9O4ggcd7XjuvYfyT56G+4nN0N2Vhh3O\nDtqMGnSv13Sb7vEaQaqsDN+G/2Lg8i/G//w1Mxm8qBHdY/4ao+5bOwvPf91rea01murzwfCQ8bgZ\nNWgzZznep8Hzl8T8Hqn7O6HNIlh27DsSLB1ex83UjGgJvkKI5LCZ3ivY/Q/KVy6nquFYys85w2hJ\nI5wJBCh6+qm4H6Z2dlL8+CbUgD/2nf0BSh55yFGSldrWeqToyu3fRbMJppHR90jWc9GWx2MmOmkz\namC2eRGO6GDpuLJVvBnRaSLBVwiRFLFKKoaLdRTsaKb8vE/m/fVgdX8nisV7oPQHIiO8hEtVWmSr\nm96196Dj+44tuuLqaLe8r+7xGve3y3oey+OBT3/adFMkWMZ5Hddpn+Z0kuArhEgKbUaN4yk89653\nqTjzY5O+3GMyOZ0OjVWqUncXWLxA7CnkI/dNPLUr3AXLnPkV55iJTj/6kW2wjPs6biL5CykmwVcI\nkRweD4MXxSgbOEIBXPv25Xe5SqfToTb3UwAlOMzw3OONEo8uF1rplMg2x+IYJTuleb0o/f3mLxcr\n0SlGsEz4Om4c+QupJsFXCJE0/rXrGF5wUlyPyedylU6nQ/1NawlceZVlNrI6MGiUeHxmG/rUqXHv\nh66Yh2odhVBtnf11Xa95Z6OBz19uWb3LNECaZS1bBcsMvY4bDwm+QojkcbvpffI5AldeheYgyxby\nvFyl0+lQt5v+r30dLGoihUs8UlyManMN1oo6kuWslU45chJw5VX0vPg3eu9/GMUmQW5w2cXmJxBr\n1zkLkCZZy3zzmzEvR2Tiddx4SIUrB7K1ek685Dhzy6QfZ18fpTddT9Gf/ogy8ofUbHw1rupRAib9\nWNMhEKD67NPhgw/GbYq8h0DF4oW4Wvcm9BKh2np6738Ybc5RRz6PQMDyObXSKXS/sRPKyswrTJlV\nsDr7HAYv+QzBk06Gyirr6lNXXI1/3Y9j77RNZavJJhWuhBDpV1aG75fr6X6vjZ4nnmXgokbTu2XL\nNMPKEiIAABIKSURBVOGkc5IFbDMd64Ta2Q7Fxcb/h6eAbZ5z4LIVRuAd2b9RU8SBAGrrXvw332aM\n7J99Ec3rpeT3v6X885+hasFcys8+g6KNj5o+d8lvN+C96frYCXkZdB03HpOX6iWEyA8eD9pHP4Zv\nw+/Qx46CljY6L1eZoaObtPrRjwj0D9m+h+H/H3WfCy+C4DAl//UbFJtCGuEmCUVPPYG6rw2tto7B\npcvwr/7uyHNuRG1tBZcKoRBFWx6HAjf+G29B7e4yPp/CwpHR7sYjz3HhUor++DCu6MpVoRAFLc2W\nFbiUUAjPhvXgLsi4WuLJINPODuTFlBZynLkmY48znkAambbcODoYNK0ddV00Y481ySLH6eQ9NLmP\nd9UNps0ywqyaJASuuJr+a/+Fkl/8DM9v7xu3XSstRQkE0Grr0KZOddxoAYzFSHaZ2cm4LDFZ7Kad\nJfg6kHe/2DlOjjN7WF4PXHkFg5+5lGDDfKisyoljdWLCxxl9MhM1gtXqZjO4ZAlFW57A1Tq+3rTu\nchlrgVXVduSciFjBV3e56HnxNWNqOctI8J0g+cXOLXKcWcIm0SfyR8vlIjivgYJXX+HA4eG07t5k\nSNpnGh4Vl5Wh9vWhzahB3d9JxaJTM66VYa6OfCXhSgiRkeyqGCnhf6GQMcW5aFFa9y3rhZOUKqsi\nyUp2hSsmU64m5EnwFUJkpLiCwZtv4t74v9I1aSImmCmdDDoQmjkLsnDdbrwk21kIkZlGgoHZNd9x\ndJ3yK74Eqkpw9lH0/s9jkIGjuEw3NlMaML3Gq7tcoOvoHi+qz9k0+Njrm2bXebX6OfQ8+SzVBRo9\n7tKcHPGGychXCJGxRlUxUlXbxvAKoGgaBR/spuqjDZSfe2Z2tS6M0RQ+LcZU3Or/yhWmd+v/8pX0\nvPR3ut/YOa7KlFV5UWXMPzODSxuhsgqOPTb1gXeS328JvkKIzBUdDF76u5Hd7IACFDRvp7zxvNTu\nXzI4bAqfViPXhP1r7rAsHakdfQyUlY0rj9n75HNHHuPw5XSMRgxoWuqPO0Peb8l2diDrs0YdkuPM\nLTl5nAMDlDeeh3tnC4xMh8ZaptL16luow8MZW6DDcjnVNdeNKy4xaZ9pIkVOuruo+NTpuA58GN9L\nXXMdnrt+mbLjjOf9nijJdhZC5IbiYnqf2UZX8y567/1d7FZ4oRAVF56TOSPKseJsCj9pEijhqPb1\noXYdiPulUnrcGfR+S/AVQmSfyiqCy/4p5jS0ArgOfIiiabha92Zc/+C4m8JnEa2gAK2qKu7Hqe1t\n0NGRgj3KrPdbgq8QImv1bno6q/sHJ9wUPpMNDFB+7plUnXYy6gHzka/O+OznMG1WHcycmZJdy6T3\nW4KvECL9kpVpOjINTWsrw3OOQlcU4w+7TWZ0Ro0oc6Ap/FjljedR0LwdJRSyvB4fM+M5VcedQe+3\nrPMVQqSPw0YJcauro/fVt6C7C3fLDoJHH0PFp5ea96DNsBGlaRcip92eMk13l5EM55BWOgV9ajlq\nZ/uo405lCMyU91uCrxAibbxNt4zKNA1fh1UOHcK37icTH3lUVhFcfBaAZYGOjBtRjiyn8t98W9a3\nTXS37IhkoTuh9Ac4uHELFBen77gz5P2W4CuESA+bTNPiB++n8IXnGVx2ycRHwSMyZYTjWDijOIsF\nG+aDy+U4AGuz6tDmHDU5JxuT/H5L8BVCpEWsRgmufW3JHQVnyAgnr1RWGV2mHPbzzbhZiDSShCsh\nRFo4bZRQ/OD9VJz58eStyU1gjWpEJpR8zDK9m55muGGBZcKbDoTq6nO6aYITEnyFEOnhsGtO9Ch4\n0tbkmpUgvOl61Pd2SSCOpbiYvnt/Z71dVen9/UNGNakkXF7IVhJ8hRBpE2mUUFfvqO7vZK3JDSeG\nuVr3HinQsWE9FWd8LDMrZWUYbUYNWv1s82219Whzjk7zHmUeCb5CiPQJN0p44VUGVlwe8+6TsibX\nJjFMgYyslJVxMmg9baaS4CuESD+PB99P/iPmKHgy1uTaJYZFy6RKWZloVDvI6I5IeXydN5oEXyHE\n5HAwCp6MUZLTxLDIqDwQQN3ZgrpzR/KCcS4keo3pDdyz9eXMuM6bIe+tBF8hxOSKHgVnwijJYWKY\nNrOWkl/9nMoFx1Fx1ulUnLWIyvnHUfr1a6CvL7HXzpBes6YSDVoTyTZPpgx7bxPq53v48GG+/e1v\n4/P5GB4e5qabbuKjH/2o7WOyua9oTvZFNSHHmVuy8jgT6RvLyLHu2Z+89byRMpibUFv3mNYhHl5w\nkuV6Vq20lIHLV8ZdMMSq12z/isvx/eAnVM+Zkf7PNFUlQW2k4rubzj6+YUnv53vvvfdy+umn87vf\n/Y7vf//7fO9730t454QQIiKRUVIwCN/8ZnJHNNFTptv+RuCKq0ePyq+8CvXgQcuHqz5f/ElZdhXA\nHrifik+eBt/8ZtpHaqaZ39mWcJZBfXzDEjpt+epXv0phYSEAoVCIoqKipO6UEEI45W26Be7+Fa6R\nn8PBAZj4iMbjQZt7PP51P8YfNSpX93dSct+GmA8v2rwJ/823OTqZiFkBrK0V7rwTb/9QykZq48QI\nWk6PbbI56eOb7lKTMaedH374YX7zm9+Muu3222/n5JNP5sCBA1x99dXcfPPNLFy40PaFgsEQbrfL\n9j5CCBGXQAAaGmDPnvHbjjoKduyIPzgEAkYz95kzrR8bCMC8ebB3fNekUVwueOcdOPZYZ687fz58\n8IH9/RI9rkT84x9w/PGgaeO3xXNsk83uvU3n+xkl5sj3sssu47LLLht3+zvvvMP111/PjTfeGDPw\nAhw8mL1Ze1l57SwBcpy5JR+OU31/NxWtrabXZPXWVnqadzkf0cR5bdN7kXnXpGihWXX0uEsh/DnE\nuKbtXbI05nOOOq4Er5E75i6lorbOtDXjuGNLopRc87V4bwNLluL3h8CfmuOwktC083vvvcc3vvEN\nfvazn3HiiScmvGNCCDER4WVByejba9XuEMynr/1Na0HTKH7gfhTfYdMTgMhSKYeBPdKJaeNjqPvM\nTyq0WXVolVV4V69KfRLUSOZ3VrRmjCHTulwllO183XXX8c4771BbWwtAaWkpv/qV/dlaNp+B58MI\nAuQ4c02+HGdSslgDASoWLzQf4dXPoWfry7ZT0Oo/3sPzy/9L4csvoXaMbgyP2x3/PgYClK66npIH\n7zd9DJC+zN3ozO+xQStTsp3jmQFI9WxBFLuRb0LBNxHZ/EcgX/6IyXHmlnw5ToJBqtd9l9Af/5Rw\ncFDf303FolNRTK5t6i4XPS++5mz62uwPe6zA/udNuN9uQausRjvhxCOPMwl6rks/w4F//TYV55yR\n2InCRKQ5aDn67k7CMqh4SPCdoHz5IybHmVvy5ThhAut8wwGlrIyKJWenJKDZBvaxP7vdDH7ms/h+\n8GMoKxu9jzNqqJ4zg+5X3kzOiUIGc/rdnYy1u/FI+jpfIYTIOPGsER5b7WjJ2WhTp5redaLXNu3K\nVSpj/qnBICWPPEjlKSceWas85rjsnm8yamFPmgxcuxsPCb5CiLxjVjiioHk7wwtOSn6JS4flKqPZ\nFumQjkGAs7W7mWzyJ8WFECKdbEZM6qE+4xrs+7sJNsyHyqqkvOS4TNuKStQDH5pmM0ezKmSRaZm7\nkyGZme6TQYKvECKv2I6YWvdSsex81P37k5u8M1Ku0n/zbcb124ICqj5+Mmgh+321qr409vnSkASV\ncbJ8GZQEXyFEXrEbMSnouDo6gCSXqQwLX78Fgg0Nlo0ZIvs6dgQ30lkoEmyjni8fZfMMgFzzFULk\nlzivwaYqead309MMz1+Arijjsp7DRhXpWL0K5s/PiHZ4GSNTewY7kPl7KIQQSTZuxDR9BmpHu+k1\n2JQV3i8upvfZF6G7C/ffXqX4kYco/NsrqJ3t40Zw0dW3FFI0Ks9mWTgDIOt8HciX9ZJynLklX44T\nJnCsE1nnm4qiE4kU6UhVUY1JlCvfXVnnK4QQZsIjpsoq58t3xq4RTuYUsMla5WxfUiPMybSzEELg\nPHkn3gYME5XtS2qEORn5CiEEOEvemYyqSlJUIyfJyFcIIaLZJO84mQLWZtQk/VpwePTteXIzemtr\nVi2pEeYk+AohhEO2U8Azayn59S8oeuqJ5HfYGRmVe376I3qad+VnUY0cI9POQgjhlM0UsFY+Fc+9\n94yqF+25+1eU/tvXkzcdHU/zCJHRJPgKIUQc/E1rCVxz3egGDFdehXqo1/T+xQ/eT8WZH5eiGGIU\nmXYWQoh4mNRVVvd3UnLfBtO7K4BrX5sUxRCjyMhXCCESETUFbNdjN1o29JkV6SHBVwghJsphvWgp\niiHCZNpZCCGSIFKkY9NjqG2tpnWiLYtipKJUpchoMvIVQohkCBfpeOFVBlZcbnqXtJaqFBlNRr5C\nCJFMHg++n/wHetnUjCtVKTKHBF8hhEg2k4xos65IdqUq/TffJlPQOUymnYUQIlVsimJIt6L8JsFX\nCCEmgd3yJOlWFIdAAPX93Vm3hEuCrxBCTAbpVjQxWZ6sJtd8hRBikjjtISzGy/ZkNQm+QggxWZwk\nZonxciBZTaadhRBiskm3orjkQrKaBF8hhBBZJReS1ST4CiGEyC45kKwm13yFEEJknWxPVpPgK4QQ\nIvtkebKaBF8hhBDZK5yslmXkmq8QQgiRZhJ8hRBCiDST4CuEEEKkmQRfIYQQIs0k+AohhBBpJsFX\nCCGESDMJvkIIIUSaSfAVQggh0kyCrxBCCJFmiq7r+mTvhBBCCJFPZOQrhBBCpJkEXyGEECLNJPgK\nIYQQaSbBVwghhEgzCb5CCCFEmknwFUIIIdJMgu8ITdO49dZbWb58OStXrmTPnj2jtj/zzDN89rOf\nZfny5Tz00EOTtJcTF+s477vvPpYtW8bKlStZuXIlu3fvnqQ9TY4333yTlStXjrs9Vz7PaFbHmiuf\n6fDwMN/+9re5/PLL+dznPsfTTz89anuufKaxjjNXPs9QKMR3vvMdVqxYwRe+8AXefffdUdtz5fO0\npAtd13X9iSee0FetWqXruq6//vrr+rXXXhvZNjQ0pJ9//vl6b2+vPjg4qF966aX6gQMHJmtXJ8Tu\nOHVd12+44QZ9+/btk7FrSXf33XfrF198sX7ZZZeNuj2XPs8wq2PV9dz5TB955BF9zZo1uq7r+sGD\nB/Wzzjorsi2XPlO749T13Pk8t2zZot900026ruv6X//6/9q7d5dGwiiMwz9NGJCMCjZ2aQQrCy+V\nhaTxBioWKipCBO29FOIFBCEgWAtB7fwPBCuxEARhRQMWamdho42FyEwECc4WGwdcnQTWOGE/36eb\nOc05vMl3yNfkl7FnbhD98s3LZDJ0dHQA0NzczOXlpV+7ubkhHo9TW1uLZVm0tbVxdnZWrla/pNCc\nAFdXV+zs7DA+Ps729nY5WiyZeDzO5ubmh/cm5fkmaFYwJ9Pe3l5mZ2cB8DyPSCTi10zKtNCcYE6e\nnZ2dpFIpAO7u7qipqfFrJuUZRMs3z3EcbNv2nyORCLlczq9VV1f7tVgshuM4ofdYCoXmBOjr62Nt\nbY3d3V0ymQxHR0flaLMkenp6iEajH96blOeboFnBnExjsRi2beM4DjMzM8zNzfk1kzItNCeYkydA\nNBplcXGRVCrFwMCA/96kPINo+ebZto3ruv7z6+urf5j9XXNd990H439SaE7P85icnKSurg7Lskgk\nElxfX5er1W9jUp7FmJbp/f09yWSSwcHBd4e1aZkGzWlangAbGxscHBywurpKNpsFzMvzM1q+ea2t\nrRwfHwNwcXFBY2OjX2toaOD29pbHx0deXl44Pz+npaWlXK1+SaE5Hcehv78f13XxPI/T01OamprK\n1eq3MSnPYkzK9OHhgampKRYWFhgeHn5XMynTQnOalOfe3p5/bV5VVUVFRQWVlX9Wkkl5Bvn8nuoH\n6urq4uTkhLGxMTzPY319nf39fbLZLKOjoywtLTE9PY3neQwNDVFfX1/ulv9JsTnn5+dJJpNYlkV7\nezuJRKLcLZeMiXkGMTHTra0tnp6eSKfTpNNpAEZGRnh+fjYq02JzmpJnd3c3y8vLTExMkMvlWFlZ\n4fDw8Md8R/WvRiIiIiHTtbOIiEjItHxFRERCpuUrIiISMi1fERGRkGn5ioiIhEzLV0REJGRaviIi\nIiHT8hUREQnZb7rvKdyQrPs/AAAAAElFTkSuQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "y_pred = tree.predict(X)\n", "show_predicitons(X, y, y_pred)" ] }, { "cell_type": "code", "execution_count": 373, "metadata": {}, "outputs": [], "source": [ "#!pip install modelvis" ] }, { "cell_type": "code", "execution_count": 374, "metadata": {}, "outputs": [], "source": [ "import modelvis" ] }, { "cell_type": "code", "execution_count": 375, "metadata": {}, "outputs": [ { "data": { "image/svg+xml": [ "\n", "\n", "Tree\n", "\n", "\n", "0\n", "\n", "distance_log ≤ 1.9442\n", "mse = 0.3069\n", "samples = 1464\n", "value = -0.692\n", "\n", "\n", "1\n", "\n", "distance_log ≤ 1.3756\n", "mse = 0.2053\n", "samples = 582\n", "value = -0.2194\n", "\n", "\n", "0->1\n", "\n", "\n", "True\n", "\n", "\n", "4\n", "\n", "distance_log ≤ 2.2855\n", "mse = 0.1293\n", "samples = 882\n", "value = -1.0038\n", "\n", "\n", "0->4\n", "\n", "\n", "False\n", "\n", "\n", "2\n", "\n", "mse = 0.1657\n", "samples = 148\n", "value = 0.2985\n", "\n", "\n", "1->2\n", "\n", "\n", "\n", "\n", "3\n", "\n", "mse = 0.0961\n", "samples = 434\n", "value = -0.396\n", "\n", "\n", "1->3\n", "\n", "\n", "\n", "\n", "5\n", "\n", "mse = 0.0882\n", "samples = 405\n", "value = -0.7747\n", "\n", "\n", "4->5\n", "\n", "\n", "\n", "\n", "6\n", "\n", "mse = 0.0818\n", "samples = 477\n", "value = -1.1983\n", "\n", "\n", "4->6\n", "\n", "\n", "\n", "\n", "" ], "text/plain": [ "" ] }, "execution_count": 375, "metadata": {}, "output_type": "execute_result" } ], "source": [ "modelvis.render_tree(tree, feature_names=['distance_log'])" ] }, { "cell_type": "code", "execution_count": 377, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([-0.11421984])" ] }, "execution_count": 377, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.predict([[1.4]])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Grid Search" ] }, { "cell_type": "code", "execution_count": 378, "metadata": {}, "outputs": [], "source": [ "from sklearn.model_selection import GridSearchCV" ] }, { "cell_type": "code", "execution_count": 380, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "DecisionTreeRegressor(criterion='mse', max_depth=4, max_features=None,\n", " max_leaf_nodes=None, min_impurity_split=1e-07,\n", " min_samples_leaf=1, min_samples_split=2,\n", " min_weight_fraction_leaf=0.0, presort=False, random_state=None,\n", " splitter='best')" ] }, "execution_count": 380, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tree = DecisionTreeRegressor()\n", "param_grid = {\n", " \"max_depth\": [2, 3, 4, 5, 6, 7, 8, 9, 10]\n", "}\n", "grid = GridSearchCV(tree, \n", " param_grid=param_grid, \n", " cv=10, \n", " scoring=\"neg_mean_squared_error\")\n", "grid.fit(X, y)\n", "grid.best_estimator_" ] }, { "cell_type": "code", "execution_count": 396, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAd8AAAFJCAYAAADaPycGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "tree = grid.best_estimator_\n", "\n", "df2 = newdata.sort_values(by='distance_log')\n", "X = df2[['distance_log']]\n", "y = df2.cartage_per_distance_log\n", "\n", "def show_predicitons2(model, X, y):\n", " X = X.as_matrix()\n", " X1 = np.arange(X[0].min(), X[0].max(), 1000)\n", " y1 = model.predict(X1)\n", " \n", "def show_predicitons(X, y, y_pred): \n", " plt.scatter(X, y, color='r')\n", " plt.plot(X, y_pred, color='b')\n", "\n", "y_pred = tree.predict(X)\n", "show_predicitons(X, y, y_pred)\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Cars dataset" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Download the [cars dataset](cars_small.csv) and build a model to predict the price of a car.\n", "\n", "features: brand, kmpl, bhp, type \n", "target: price\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.1" } }, "nbformat": 4, "nbformat_minor": 2 }