From fcffe425bc5579cb1820ec6bb6cd65a58eca57f4 Mon Sep 17 00:00:00 2001 From: nimrod-d Date: Wed, 3 Apr 2024 16:47:16 +0200 Subject: [PATCH 1/2] tech challenge --- fuel_efficency_ironhack.ipynb | 1969 +++++++++++++++++++++++++++++++++ 1 file changed, 1969 insertions(+) create mode 100644 fuel_efficency_ironhack.ipynb diff --git a/fuel_efficency_ironhack.ipynb b/fuel_efficency_ironhack.ipynb new file mode 100644 index 0000000..106d67d --- /dev/null +++ b/fuel_efficency_ironhack.ipynb @@ -0,0 +1,1969 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 529, + "id": "8ffe9c0f-678c-4b43-9f3b-65c8e0fe0309", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import matplotlib.pyplot as plt \n", + "from imblearn.under_sampling import RandomUnderSampler\n", + "from scipy import stats\n", + "from bs4 import BeautifulSoup\n", + "import requests" + ] + }, + { + "cell_type": "code", + "execution_count": 530, + "id": "8c97e2b7-19c9-491f-b13b-2368ea9fb236", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "url1 = 'measurements.csv'\n", + "url2 = 'measurements2.xlsx'\n", + "\n", + "df1 = pd.read_csv(url1)\n", + "df2 = pd.read_excel(url2)" + ] + }, + { + "cell_type": "markdown", + "id": "fcb5a5bb-ed6e-42d5-98e5-804c0593f5bb", + "metadata": {}, + "source": [ + "# Cleaning process: \n", + "\n", + "\n", + "\n", + "## Cleaning of df1 " + ] + }, + { + "cell_type": "code", + "execution_count": 531, + "id": "5dd40ea0-4def-40aa-bc01-9b4ae0ea07a6", + "metadata": { + "tags": [] + }, + "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", + "
distanceconsumespeedtemp_insidetemp_outsidespecialsgas_typeACrainsunrefill litersrefill gas
02852621,512NaNE1000045E10
1124,23021,513NaNE10000NaNNaN
211,25,53821,515NaNE10000NaNNaN
312,93,93621,514NaNE10000NaNNaN
418,54,54621,515NaNE10000NaNNaN
.......................................
383163,73924,518NaNSP98000NaNNaN
38416,14,3382531ACSP98100NaNNaN
385163,8452519NaNSP98000NaNNaN
38615,44,6422531ACSP98100NaNNaN
38714,75252530ACSP98100NaNNaN
\n", + "

388 rows × 12 columns

\n", + "
" + ], + "text/plain": [ + " distance consume speed temp_inside temp_outside specials gas_type AC \\\n", + "0 28 5 26 21,5 12 NaN E10 0 \n", + "1 12 4,2 30 21,5 13 NaN E10 0 \n", + "2 11,2 5,5 38 21,5 15 NaN E10 0 \n", + "3 12,9 3,9 36 21,5 14 NaN E10 0 \n", + "4 18,5 4,5 46 21,5 15 NaN E10 0 \n", + ".. ... ... ... ... ... ... ... .. \n", + "383 16 3,7 39 24,5 18 NaN SP98 0 \n", + "384 16,1 4,3 38 25 31 AC SP98 1 \n", + "385 16 3,8 45 25 19 NaN SP98 0 \n", + "386 15,4 4,6 42 25 31 AC SP98 1 \n", + "387 14,7 5 25 25 30 AC SP98 1 \n", + "\n", + " rain sun refill liters refill gas \n", + "0 0 0 45 E10 \n", + "1 0 0 NaN NaN \n", + "2 0 0 NaN NaN \n", + "3 0 0 NaN NaN \n", + "4 0 0 NaN NaN \n", + ".. ... ... ... ... \n", + "383 0 0 NaN NaN \n", + "384 0 0 NaN NaN \n", + "385 0 0 NaN NaN \n", + "386 0 0 NaN NaN \n", + "387 0 0 NaN NaN \n", + "\n", + "[388 rows x 12 columns]" + ] + }, + "execution_count": 531, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df1" + ] + }, + { + "cell_type": "code", + "execution_count": 532, + "id": "e18d76f5-e744-4c12-a5b0-39a96b0413ea", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "SP98 228\n", + "E10 160\n", + "Name: gas_type, dtype: int64" + ] + }, + "execution_count": 532, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df1['gas_type'].value_counts()" + ] + }, + { + "cell_type": "code", + "execution_count": 533, + "id": "8530ad4a-278b-4741-85d4-25baef458b14", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "SP98 228\n", + "E10 160\n", + "Name: gas_type, dtype: int64" + ] + }, + "execution_count": 533, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df2['gas_type'].value_counts()" + ] + }, + { + "cell_type": "code", + "execution_count": 534, + "id": "2c36911f-405d-473f-bf61-d00c286ffbb8", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "rows_to_remove1 = df1[df1['gas_type'] == 'SP98'].sample(n=68, random_state=42).index\n", + "rows_to_remove2 = df2[df2['gas_type'] == 'SP98'].sample(n=68, random_state=42).index" + ] + }, + { + "cell_type": "code", + "execution_count": 535, + "id": "94fd800d-3cef-404c-a65e-f78aa5fa7438", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df1 = df1.drop(rows_to_remove1, axis=0)\n", + "df2 = df2.drop(rows_to_remove2, axis=0)" + ] + }, + { + "cell_type": "code", + "execution_count": 536, + "id": "e809c8bd-99f0-43a7-aaea-3bcdaa7b9f33", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "E10 160\n", + "SP98 160\n", + "Name: gas_type, dtype: int64\n", + "E10 160\n", + "SP98 160\n", + "Name: gas_type, dtype: int64\n" + ] + } + ], + "source": [ + "print(df1['gas_type'].value_counts())\n", + "print(df2['gas_type'].value_counts())" + ] + }, + { + "cell_type": "code", + "execution_count": 537, + "id": "ccf72746-da28-486b-ba43-598f73df87d4", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Int64Index: 320 entries, 0 to 387\n", + "Data columns (total 12 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 distance 320 non-null object\n", + " 1 consume 320 non-null object\n", + " 2 speed 320 non-null int64 \n", + " 3 temp_inside 309 non-null object\n", + " 4 temp_outside 320 non-null int64 \n", + " 5 specials 73 non-null object\n", + " 6 gas_type 320 non-null object\n", + " 7 AC 320 non-null int64 \n", + " 8 rain 320 non-null int64 \n", + " 9 sun 320 non-null int64 \n", + " 10 refill liters 9 non-null object\n", + " 11 refill gas 9 non-null object\n", + "dtypes: int64(5), object(7)\n", + "memory usage: 32.5+ KB\n" + ] + } + ], + "source": [ + "df1.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 538, + "id": "c2a75aed-d884-4902-b243-c348a3e04ac3", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "columns_to_numeric = ['distance', 'consume', 'temp_inside', 'refill liters']\n", + "\n", + "for column in columns_to_numeric:\n", + " df1[column] = df1[column].fillna(\"0\")\n", + " df1[column] = df1[column].replace(\",\", \".\", regex=True)\n", + " df1[column] = df1[column].astype(float)\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 539, + "id": "2ed7bd03-a42d-4293-96ea-a00963d5f4f8", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df1_dummies = pd.get_dummies(df1, columns=['gas_type'])\n" + ] + }, + { + "cell_type": "code", + "execution_count": 540, + "id": "b7a59ba4-e4a1-4abb-93f1-39b01f1b6734", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "# df1_dummies.drop('specials', axis=1, inplace=True)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 541, + "id": "75583173-ea55-4e41-bb96-39a1cd34f0f0", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df1_ready = df1_dummies " + ] + }, + { + "cell_type": "code", + "execution_count": 542, + "id": "e5d475ff-282f-4384-ac7a-728529242aba", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Int64Index: 320 entries, 0 to 387\n", + "Data columns (total 13 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 distance 320 non-null float64\n", + " 1 consume 320 non-null float64\n", + " 2 speed 320 non-null int64 \n", + " 3 temp_inside 320 non-null float64\n", + " 4 temp_outside 320 non-null int64 \n", + " 5 specials 73 non-null object \n", + " 6 AC 320 non-null int64 \n", + " 7 rain 320 non-null int64 \n", + " 8 sun 320 non-null int64 \n", + " 9 refill liters 320 non-null float64\n", + " 10 refill gas 9 non-null object \n", + " 11 gas_type_E10 320 non-null uint8 \n", + " 12 gas_type_SP98 320 non-null uint8 \n", + "dtypes: float64(4), int64(5), object(2), uint8(2)\n", + "memory usage: 30.6+ KB\n" + ] + } + ], + "source": [ + "df1_ready.info()" + ] + }, + { + "cell_type": "markdown", + "id": "39f57bd8-5f1f-45c0-8364-585257864e69", + "metadata": {}, + "source": [ + "## Cleaned df1 " + ] + }, + { + "cell_type": "code", + "execution_count": 543, + "id": "3dacb65b-9840-4faa-9e4d-3080ac810cdf", + "metadata": { + "tags": [] + }, + "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", + "
distanceconsumespeedtemp_insidetemp_outsidespecialsACrainsunrefill litersrefill gasgas_type_E10gas_type_SP98
028.05.02621.512NaN00045.0E1010
112.04.23021.513NaN0000.0NaN10
211.25.53821.515NaN0000.0NaN10
312.93.93621.514NaN0000.0NaN10
418.54.54621.515NaN0000.0NaN10
\n", + "
" + ], + "text/plain": [ + " distance consume speed temp_inside temp_outside specials AC rain \\\n", + "0 28.0 5.0 26 21.5 12 NaN 0 0 \n", + "1 12.0 4.2 30 21.5 13 NaN 0 0 \n", + "2 11.2 5.5 38 21.5 15 NaN 0 0 \n", + "3 12.9 3.9 36 21.5 14 NaN 0 0 \n", + "4 18.5 4.5 46 21.5 15 NaN 0 0 \n", + "\n", + " sun refill liters refill gas gas_type_E10 gas_type_SP98 \n", + "0 0 45.0 E10 1 0 \n", + "1 0 0.0 NaN 1 0 \n", + "2 0 0.0 NaN 1 0 \n", + "3 0 0.0 NaN 1 0 \n", + "4 0 0.0 NaN 1 0 " + ] + }, + "execution_count": 543, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df1_ready.head()" + ] + }, + { + "cell_type": "markdown", + "id": "cce1099c-db53-44b0-a5ec-a8655487b091", + "metadata": {}, + "source": [ + "## Cleaning of df2" + ] + }, + { + "cell_type": "code", + "execution_count": 544, + "id": "2f0c08e3-06b3-47c0-ad39-fe516504b2f1", + "metadata": { + "tags": [] + }, + "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", + "
distanceconsumespeedtemp_insidetemp_outsidespecialsgas_typeACrainsunrefill litersrefill gas
028.05.02621.512NaNE1000045.0E10
112.04.23021.513NaNE10000NaNNaN
211.25.53821.515NaNE10000NaNNaN
312.93.93621.514NaNE10000NaNNaN
418.54.54621.515NaNE10000NaNNaN
\n", + "
" + ], + "text/plain": [ + " distance consume speed temp_inside temp_outside specials gas_type AC \\\n", + "0 28.0 5.0 26 21.5 12 NaN E10 0 \n", + "1 12.0 4.2 30 21.5 13 NaN E10 0 \n", + "2 11.2 5.5 38 21.5 15 NaN E10 0 \n", + "3 12.9 3.9 36 21.5 14 NaN E10 0 \n", + "4 18.5 4.5 46 21.5 15 NaN E10 0 \n", + "\n", + " rain sun refill liters refill gas \n", + "0 0 0 45.0 E10 \n", + "1 0 0 NaN NaN \n", + "2 0 0 NaN NaN \n", + "3 0 0 NaN NaN \n", + "4 0 0 NaN NaN " + ] + }, + "execution_count": 544, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df2.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 545, + "id": "e470bc44-8a6a-4402-8333-e743f138c8f5", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Int64Index: 320 entries, 0 to 387\n", + "Data columns (total 12 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 distance 320 non-null float64\n", + " 1 consume 320 non-null float64\n", + " 2 speed 320 non-null int64 \n", + " 3 temp_inside 309 non-null float64\n", + " 4 temp_outside 320 non-null int64 \n", + " 5 specials 73 non-null object \n", + " 6 gas_type 320 non-null object \n", + " 7 AC 320 non-null int64 \n", + " 8 rain 320 non-null int64 \n", + " 9 sun 320 non-null int64 \n", + " 10 refill liters 9 non-null float64\n", + " 11 refill gas 9 non-null object \n", + "dtypes: float64(4), int64(5), object(3)\n", + "memory usage: 32.5+ KB\n" + ] + } + ], + "source": [ + "df2.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 546, + "id": "73a8d3df-b7e0-4fb8-9a44-64ff57c451ab", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df2.drop('specials', axis=1, inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 547, + "id": "593848f3-4c3b-4755-8587-82633b05827b", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df2['refill liters'] = df2['refill liters'].fillna(0)" + ] + }, + { + "cell_type": "code", + "execution_count": 548, + "id": "574729f4-7149-4a73-ab5d-103f53714826", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df2 = pd.get_dummies(df2, columns=['gas_type'])" + ] + }, + { + "cell_type": "markdown", + "id": "f5575e4b-5cce-4d79-9e39-528f10de25b2", + "metadata": {}, + "source": [ + "## Cleaned df2" + ] + }, + { + "cell_type": "code", + "execution_count": 549, + "id": "5f703694-42da-4c6d-b6cc-8f2c29575796", + "metadata": { + "tags": [] + }, + "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", + "
distanceconsumespeedtemp_insidetemp_outsideACrainsunrefill litersrefill gasgas_type_E10gas_type_SP98
028.05.02621.51200045.0E1010
112.04.23021.5130000.0NaN10
211.25.53821.5150000.0NaN10
312.93.93621.5140000.0NaN10
418.54.54621.5150000.0NaN10
\n", + "
" + ], + "text/plain": [ + " distance consume speed temp_inside temp_outside AC rain sun \\\n", + "0 28.0 5.0 26 21.5 12 0 0 0 \n", + "1 12.0 4.2 30 21.5 13 0 0 0 \n", + "2 11.2 5.5 38 21.5 15 0 0 0 \n", + "3 12.9 3.9 36 21.5 14 0 0 0 \n", + "4 18.5 4.5 46 21.5 15 0 0 0 \n", + "\n", + " refill liters refill gas gas_type_E10 gas_type_SP98 \n", + "0 45.0 E10 1 0 \n", + "1 0.0 NaN 1 0 \n", + "2 0.0 NaN 1 0 \n", + "3 0.0 NaN 1 0 \n", + "4 0.0 NaN 1 0 " + ] + }, + "execution_count": 549, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df2.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 550, + "id": "c2c35b03-cd34-4762-b7ee-055386c8de2a", + "metadata": { + "tags": [] + }, + "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", + "
distanceconsumespeedtemp_insidetemp_outsidespecialsACrainsunrefill litersrefill gasgas_type_E10gas_type_SP98
028.05.02621.512NaN00045.0E1010
112.04.23021.513NaN0000.0NaN10
211.25.53821.515NaN0000.0NaN10
312.93.93621.514NaN0000.0NaN10
418.54.54621.515NaN0000.0NaN10
\n", + "
" + ], + "text/plain": [ + " distance consume speed temp_inside temp_outside specials AC rain \\\n", + "0 28.0 5.0 26 21.5 12 NaN 0 0 \n", + "1 12.0 4.2 30 21.5 13 NaN 0 0 \n", + "2 11.2 5.5 38 21.5 15 NaN 0 0 \n", + "3 12.9 3.9 36 21.5 14 NaN 0 0 \n", + "4 18.5 4.5 46 21.5 15 NaN 0 0 \n", + "\n", + " sun refill liters refill gas gas_type_E10 gas_type_SP98 \n", + "0 0 45.0 E10 1 0 \n", + "1 0 0.0 NaN 1 0 \n", + "2 0 0.0 NaN 1 0 \n", + "3 0 0.0 NaN 1 0 \n", + "4 0 0.0 NaN 1 0 " + ] + }, + "execution_count": 550, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df1_ready.head()" + ] + }, + { + "cell_type": "markdown", + "id": "4f9d8c7f-e6e7-4f96-a637-5809bb1430bd", + "metadata": {}, + "source": [ + "## Merged DF's" + ] + }, + { + "cell_type": "code", + "execution_count": 551, + "id": "e8b387f4-c17d-4ee4-88af-99a96bc8298a", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df = pd.concat([df1_ready, df2], axis=0)" + ] + }, + { + "cell_type": "code", + "execution_count": 552, + "id": "f6eadc4d-df44-4e3e-a2ed-1fd41fe70a08", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df['fuel_efficiency'] = df['distance'] / df['consume']" + ] + }, + { + "cell_type": "code", + "execution_count": 553, + "id": "c885817f-54b9-42a0-8a44-79582722c10d", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df['temp_diff'] = df['temp_inside'] - df['temp_outside']" + ] + }, + { + "cell_type": "code", + "execution_count": 554, + "id": "f5f1b938-17bb-409d-9a14-b1ab9573b7dd", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "df['km_per_liter'] = 100 / df['consume']" + ] + }, + { + "cell_type": "code", + "execution_count": 555, + "id": "1baa08fc-a9a9-4e5e-8789-b765c8f56ed8", + "metadata": { + "tags": [] + }, + "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", + "
countmeanstdmin25%50%75%max
distance640.020.45687524.5688371.30000011.80000015.35000019.000000216.100000
consume640.04.9493751.0483163.3000004.3000004.8000005.30000012.200000
speed640.042.90312513.95768214.00000033.00000042.00000051.00000090.000000
temp_inside629.021.5294123.0347530.00000021.50000022.00000022.50000025.500000
temp_outside640.011.1437507.049875-5.0000006.00000010.00000016.00000031.000000
AC640.00.0718750.2584830.0000000.0000000.0000000.0000001.000000
rain640.00.1218750.3273970.0000000.0000000.0000000.0000001.000000
sun640.00.0750000.2635970.0000000.0000000.0000000.0000001.000000
refill liters640.01.1212506.6146980.0000000.0000000.0000000.00000045.000000
gas_type_E10640.00.5000000.5003910.0000000.0000000.5000001.0000001.000000
gas_type_SP98640.00.5000000.5003910.0000000.0000000.5000001.0000001.000000
fuel_efficiency640.04.3393595.0706840.1130432.3349063.1076924.37155445.869565
temp_diff629.010.4022267.397520-21.0000006.00000011.50000015.50000026.000000
km_per_liter640.020.8548763.3056978.19672118.86792520.83333323.25581430.303030
\n", + "
" + ], + "text/plain": [ + " count mean std min 25% 50% \\\n", + "distance 640.0 20.456875 24.568837 1.300000 11.800000 15.350000 \n", + "consume 640.0 4.949375 1.048316 3.300000 4.300000 4.800000 \n", + "speed 640.0 42.903125 13.957682 14.000000 33.000000 42.000000 \n", + "temp_inside 629.0 21.529412 3.034753 0.000000 21.500000 22.000000 \n", + "temp_outside 640.0 11.143750 7.049875 -5.000000 6.000000 10.000000 \n", + "AC 640.0 0.071875 0.258483 0.000000 0.000000 0.000000 \n", + "rain 640.0 0.121875 0.327397 0.000000 0.000000 0.000000 \n", + "sun 640.0 0.075000 0.263597 0.000000 0.000000 0.000000 \n", + "refill liters 640.0 1.121250 6.614698 0.000000 0.000000 0.000000 \n", + "gas_type_E10 640.0 0.500000 0.500391 0.000000 0.000000 0.500000 \n", + "gas_type_SP98 640.0 0.500000 0.500391 0.000000 0.000000 0.500000 \n", + "fuel_efficiency 640.0 4.339359 5.070684 0.113043 2.334906 3.107692 \n", + "temp_diff 629.0 10.402226 7.397520 -21.000000 6.000000 11.500000 \n", + "km_per_liter 640.0 20.854876 3.305697 8.196721 18.867925 20.833333 \n", + "\n", + " 75% max \n", + "distance 19.000000 216.100000 \n", + "consume 5.300000 12.200000 \n", + "speed 51.000000 90.000000 \n", + "temp_inside 22.500000 25.500000 \n", + "temp_outside 16.000000 31.000000 \n", + "AC 0.000000 1.000000 \n", + "rain 0.000000 1.000000 \n", + "sun 0.000000 1.000000 \n", + "refill liters 0.000000 45.000000 \n", + "gas_type_E10 1.000000 1.000000 \n", + "gas_type_SP98 1.000000 1.000000 \n", + "fuel_efficiency 4.371554 45.869565 \n", + "temp_diff 15.500000 26.000000 \n", + "km_per_liter 23.255814 30.303030 " + ] + }, + "execution_count": 555, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.describe().T" + ] + }, + { + "cell_type": "code", + "execution_count": 557, + "id": "812cb37b-6c9d-4e36-8b20-d30b2fa41037", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "df.hist(figsize=(8,8), grid=False, bins=20)\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "code", + "execution_count": 560, + "id": "8a7b334d-2379-4049-9527-53ed2a7362ef", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 560, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# counts = df_b['gas_type_SP98'].value_counts()\n", + "sp98 = len(df[df['gas_type_SP98'] == 1])\n", + "e10 = len(df[df['gas_type_E10'] == 1])\n", + "\n", + "\n", + "plt.bar([\"SP98\", \"E10\"], [sp98, e10], color=['tab:red', 'tab:blue'])" + ] + }, + { + "cell_type": "code", + "execution_count": 561, + "id": "e5685a63-596f-4ecd-8dfb-6fd66a9d8cd8", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "# which gas type have a better km per liter - use hypothesis testing \n" + ] + }, + { + "cell_type": "markdown", + "id": "90f82713-0fd0-4c61-aea1-13d3c1de0969", + "metadata": {}, + "source": [ + "## Fecthing data from online source\n", + "\n", + "- Online data for different gas types and their current price in euro\n", + "\n", + "- creating a dataframe from the online data" + ] + }, + { + "cell_type": "code", + "execution_count": 562, + "id": "efc9f088-8470-4d12-bb3e-53e25741e7aa", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "200\n" + ] + } + ], + "source": [ + "request = requests.get('https://autotraveler.ru/en/germany/')\n", + "\n", + "print(request.status_code)\n", + "soup = BeautifulSoup(request.content)" + ] + }, + { + "cell_type": "code", + "execution_count": 563, + "id": "ab7b186a-0d3c-4317-b700-7ab6bf00565a", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "items = soup.find_all(\"li\", itemprop=\"priceSpecification\") " + ] + }, + { + "cell_type": "code", + "execution_count": 564, + "id": "8f66e240-4754-41b2-9aac-139bb4e08f11", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "# Creating a df from online data\n", + "\n", + "gas_types = [item.span.text for item in items]\n", + "gas_prices = [item.find(\"span\", itemprop=\"price\").text for item in items]\n", + "\n", + "gas_price = pd.Series(gas_prices)\n", + "gas_type = pd.Series(gas_types)\n", + "\n", + "gas_df = pd.DataFrame([gas_price, gas_type]).T\n", + "gas_df = gas_df.rename(columns={0:\"price\", 1:\"type\"})" + ] + }, + { + "cell_type": "code", + "execution_count": 565, + "id": "c86a20de-a643-4cf1-8ccf-57bf6cb1482e", + "metadata": { + "tags": [] + }, + "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", + "
pricetype
01.838E10
11.895Super (95)
21.966Super Plus (98)
31.742Diesel
41.059LPG
\n", + "
" + ], + "text/plain": [ + " price type\n", + "0 1.838 E10\n", + "1 1.895 Super (95)\n", + "2 1.966 Super Plus (98)\n", + "3 1.742 Diesel\n", + "4 1.059 LPG" + ] + }, + "execution_count": 565, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "gas_df" + ] + }, + { + "cell_type": "markdown", + "id": "e398b74f-40a5-4038-bdb3-0fb6987606ae", + "metadata": {}, + "source": [ + "# Analysis\n" + ] + }, + { + "cell_type": "code", + "execution_count": 568, + "id": "30665bbc-bed1-40ec-a04d-5bd0ba139dff", + "metadata": { + "tags": [] + }, + "outputs": [], + "source": [ + "\n", + "price_sp98 = float(gas_df[gas_df['type'] == 'Super Plus (98)'].values[0][0])\n", + "price_e10 = float(gas_df[gas_df['type'] == 'E10'].values[0][0])\n", + "\n", + "# cost per 100 kilometer for each gas type\n", + "cost_per_km_sp98 = sp98 * price_sp98_per_liter \n", + "cost_per_km_e10 = e10 * price_e10_per_liter\n", + "\n", + "# which gas type have a better km per liter\n", + "\n", + "avg_fuel_consumption = f_df.groupby('gas_type_E10')['consume'].mean()\n", + "\n", + "sp98 = avg_fuel_consumption[0]\n", + "e10 = avg_fuel_consumption[1]\n", + "\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 569, + "id": "a9a0c50d-ac43-482e-b128-0dc7f7a88cbc", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Cost per 100 kilometer:\n", + "--------------------\n", + "SP98: 9.61 €\n", + "E10: 9.06 €\n", + "\n", + "Difference: -0.549 €\n", + "\n", + "\n", + "Avg Liter per 100km:\n", + "--------------------\n", + "SP98: 4.89\n", + "E10: 4.93\n", + "\n", + "Difference: 0.0418 \n" + ] + } + ], + "source": [ + "print(f'Cost per 100 kilometer:\\n{\"-\"*20}\\nSP98: {cost_per_km_sp98:.3} €\\nE10: {cost_per_km_e10:.3} €\\n\\nDifference: {cost_per_km_e10 - cost_per_km_sp98:.3} €')\n", + "\n", + "print(f'\\n\\nAvg Liter per 100km:\\n{\"-\"*20}\\nSP98: {sp98:.3}\\nE10: {e10:.3}\\n\\nDifference: {e10-sp98:.3} ')" + ] + }, + { + "cell_type": "code", + "execution_count": 595, + "id": "9a09adb6-fbe0-4530-97ff-faa3f01147cf", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "R-squared: 0.07163318562141474\n", + "Mean Squared Error: 0.7616868936525083\n", + "Coefficients: {'distance': 0.002187368597816319, 'speed': -0.019899781559797502, 'rain': 0.7522042011762998, 'sun': -0.4101703396267231, 'AC': 0.23825169907583016}\n" + ] + } + ], + "source": [ + "from sklearn.model_selection import train_test_split\n", + "from sklearn.linear_model import LinearRegression\n", + "from sklearn.metrics import mean_squared_error, r2_score\n", + "from sklearn.preprocessing import StandardScaler\n", + "# Salling did not produce better results\n", + "\n", + "X = df[['distance', 'speed', 'rain', 'sun', 'AC']] \n", + "y = df['consume'] \n", + "\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", + "\n", + "model = LinearRegression()\n", + "model.fit(X_train, y_train)\n", + "\n", + "y_pred = model.predict(X_test)\n", + "\n", + "# Evaluate model\n", + "mse = mean_squared_error(y_test, y_pred)\n", + "r2 = r2_score(y_test, y_pred)\n", + "print(\"R-squared:\", r2)\n", + "print(\"Mean Squared Error:\", mse)\n", + "\n", + "# Interpret coefficients\n", + "coefficients = dict(zip(X.columns, model.coef_))\n", + "print(\"Coefficients:\", coefficients)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 597, + "id": "4f22572b-9bc0-496a-88bb-d607c6ec77bf", + "metadata": { + "tags": [] + }, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "avg_fuel_sp98 = df[df['gas_type_SP98'] == 1].groupby('speed')['consume'].mean()\n", + "avg_fuel_e10 = df[df['gas_type_SP98'] == 0].groupby('speed')['consume'].mean()\n", + "\n", + "plt.plot(avg_fuel_sp98.index, avg_fuel_sp98.values, label='SP98', color='blue')\n", + "plt.plot(avg_fuel_e10.index, avg_fuel_e10.values, label='E10', color='red')\n", + "\n", + "plt.xlabel('Speed')\n", + "plt.ylabel('Average Fuel Consumption (L/100km)')\n", + "plt.title('Average Fuel Consumption vs. Speed by Gas Type')\n", + "plt.legend()\n", + "plt.grid(True)\n", + "plt.show()\n" + ] + }, + { + "cell_type": "markdown", + "id": "d69ae398-5fb2-4ca7-9481-72e0ccf7dd1e", + "metadata": {}, + "source": [ + "## Conclusions \n", + "\n", + "judging by the results above we can see the following:\n", + "\n", + "- SP98 have a better fuel efficency with 4.88 per 100km vs the E10 consuming on average 0.04 more fuel per 100km (4.93)\n", + "\n", + "- E10 on the otherhand is more cost effiective as its cheaper than the SP98 \n", + "\n", + "- Using linear reggersion and coefficients calcualtion we can see the following: \n", + "\n", + "\n", + " For the 'distance' variable, the coefficient is approximately 0.0021. This suggests that for every unit increase in distance (e.g., 1 km), fuel consumption is expected to increase by approximately 0.00218 liters per 100 km.\n", + "\n", + "\n", + " For the 'speed' variable, the coefficient is approximately -0.0198. This indicates that for every unit increase in speed (e.g., 1 km/h), fuel consumption is expected to decrease by approximately 0.0198 liters per 100 km. This negative coefficient suggests that higher speeds lead to lower fuel consumption, which is counterintuitive and may need further investigation.\n", + "\n", + "\n", + " Rainy conditions have the highest impact on fuel consumption with 0.7522\n", + " \n", + " For the 'sun' variable, the coefficient is approximately -0.4101. This suggests that sunny weather conditions are associated with lower fuel consumption\n", + " \n", + " With AC on coefficient is 0.2382 (another counterintuitive result). \n", + " \n", + "\n", + "- We can see that both gas types performing poorly in term of average fuel consumption in low speeds and become more efficent between 29 to 65\n", + "\n", + "\n", + "Conclusion:\n", + "\n", + "The high impact of rain on fuel consumption can be connected to lower speed driving, which can be seen in the graph above, to have high impact on feul consumption.\n", + "\n", + "Best speed to fuel consumption are between 29-65 for both gas types. \n", + "\n", + "The fact that AC seems to have a positive impact on fuel consumption, a counterintative thought, can be connected to the fact that when the AC is on it is most likely a sunny day which correspond to better road conidtions hence higer speeds. This tranlste to better fuel consumption. " + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "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": "3.11.5" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} From 6b0ea914fe18bd060f591f61b95500ddac3fcddc Mon Sep 17 00:00:00 2001 From: nimrod-d Date: Tue, 7 May 2024 15:53:37 +0200 Subject: [PATCH 2/2] Update fuel_efficency_ironhack.ipynb --- fuel_efficency_ironhack.ipynb | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/fuel_efficency_ironhack.ipynb b/fuel_efficency_ironhack.ipynb index 106d67d..2e72f65 100644 --- a/fuel_efficency_ironhack.ipynb +++ b/fuel_efficency_ironhack.ipynb @@ -1939,10 +1939,16 @@ "\n", "The high impact of rain on fuel consumption can be connected to lower speed driving, which can be seen in the graph above, to have high impact on feul consumption.\n", "\n", - "Best speed to fuel consumption are between 29-65 for both gas types. \n", - "\n", - "The fact that AC seems to have a positive impact on fuel consumption, a counterintative thought, can be connected to the fact that when the AC is on it is most likely a sunny day which correspond to better road conidtions hence higer speeds. This tranlste to better fuel consumption. " + "Best speed to fuel consumption are between 29-65 for both gas types. \n" ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2305be09-1ab0-4538-9b5e-a3102d888a63", + "metadata": {}, + "outputs": [], + "source": [] } ], "metadata": {