asadsheikh/cpp-project
0
1{2 "cells": [3 {4 "cell_type": "markdown",5 "metadata": {},6 "source": [7 "## **Car Price Predictor**"8 ]9 },10 {11 "cell_type": "code",12 "execution_count": 31,13 "metadata": {},14 "outputs": [],15 "source": [16 "import pandas as pd \n",17 "import numpy as np"18 ]19 },20 {21 "cell_type": "code",22 "execution_count": 32,23 "metadata": {},24 "outputs": [25 {26 "data": {27 "text/html": [28 "<div>\n",29 "<style scoped>\n",30 " .dataframe tbody tr th:only-of-type {\n",31 " vertical-align: middle;\n",32 " }\n",33 "\n",34 " .dataframe tbody tr th {\n",35 " vertical-align: top;\n",36 " }\n",37 "\n",38 " .dataframe thead th {\n",39 " text-align: right;\n",40 " }\n",41 "</style>\n",42 "<table border=\"1\" class=\"dataframe\">\n",43 " <thead>\n",44 " <tr style=\"text-align: right;\">\n",45 " <th></th>\n",46 " <th>name</th>\n",47 " <th>company</th>\n",48 " <th>year</th>\n",49 " <th>Price</th>\n",50 " <th>kms_driven</th>\n",51 " <th>fuel_type</th>\n",52 " </tr>\n",53 " </thead>\n",54 " <tbody>\n",55 " <tr>\n",56 " <th>0</th>\n",57 " <td>Hyundai Santro Xing XO eRLX Euro III</td>\n",58 " <td>Hyundai</td>\n",59 " <td>2007</td>\n",60 " <td>80,000</td>\n",61 " <td>45,000 kms</td>\n",62 " <td>Petrol</td>\n",63 " </tr>\n",64 " <tr>\n",65 " <th>1</th>\n",66 " <td>Mahindra Jeep CL550 MDI</td>\n",67 " <td>Mahindra</td>\n",68 " <td>2006</td>\n",69 " <td>4,25,000</td>\n",70 " <td>40 kms</td>\n",71 " <td>Diesel</td>\n",72 " </tr>\n",73 " <tr>\n",74 " <th>2</th>\n",75 " <td>Maruti Suzuki Alto 800 Vxi</td>\n",76 " <td>Maruti</td>\n",77 " <td>2018</td>\n",78 " <td>Ask For Price</td>\n",79 " <td>22,000 kms</td>\n",80 " <td>Petrol</td>\n",81 " </tr>\n",82 " <tr>\n",83 " <th>3</th>\n",84 " <td>Hyundai Grand i10 Magna 1.2 Kappa VTVT</td>\n",85 " <td>Hyundai</td>\n",86 " <td>2014</td>\n",87 " <td>3,25,000</td>\n",88 " <td>28,000 kms</td>\n",89 " <td>Petrol</td>\n",90 " </tr>\n",91 " <tr>\n",92 " <th>4</th>\n",93 " <td>Ford EcoSport Titanium 1.5L TDCi</td>\n",94 " <td>Ford</td>\n",95 " <td>2014</td>\n",96 " <td>5,75,000</td>\n",97 " <td>36,000 kms</td>\n",98 " <td>Diesel</td>\n",99 " </tr>\n",100 " </tbody>\n",101 "</table>\n",102 "</div>"103 ],104 "text/plain": [105 " name company year Price \\\n",106 "0 Hyundai Santro Xing XO eRLX Euro III Hyundai 2007 80,000 \n",107 "1 Mahindra Jeep CL550 MDI Mahindra 2006 4,25,000 \n",108 "2 Maruti Suzuki Alto 800 Vxi Maruti 2018 Ask For Price \n",109 "3 Hyundai Grand i10 Magna 1.2 Kappa VTVT Hyundai 2014 3,25,000 \n",110 "4 Ford EcoSport Titanium 1.5L TDCi Ford 2014 5,75,000 \n",111 "\n",112 " kms_driven fuel_type \n",113 "0 45,000 kms Petrol \n",114 "1 40 kms Diesel \n",115 "2 22,000 kms Petrol \n",116 "3 28,000 kms Petrol \n",117 "4 36,000 kms Diesel "118 ]119 },120 "execution_count": 32,121 "metadata": {},122 "output_type": "execute_result"123 }124 ],125 "source": [126 "df = pd.read_csv('quikr_car.csv')\n",127 "df.head()"128 ]129 },130 {131 "cell_type": "code",132 "execution_count": 33,133 "metadata": {},134 "outputs": [135 {136 "data": {137 "text/plain": [138 "(892, 6)"139 ]140 },141 "execution_count": 33,142 "metadata": {},143 "output_type": "execute_result"144 }145 ],146 "source": [147 "df.shape"148 ]149 },150 {151 "cell_type": "code",152 "execution_count": 34,153 "metadata": {},154 "outputs": [155 {156 "name": "stdout",157 "output_type": "stream",158 "text": [159 "<class 'pandas.core.frame.DataFrame'>\n",160 "RangeIndex: 892 entries, 0 to 891\n",161 "Data columns (total 6 columns):\n",162 " # Column Non-Null Count Dtype \n",163 "--- ------ -------------- ----- \n",164 " 0 name 892 non-null object\n",165 " 1 company 892 non-null object\n",166 " 2 year 892 non-null object\n",167 " 3 Price 892 non-null object\n",168 " 4 kms_driven 840 non-null object\n",169 " 5 fuel_type 837 non-null object\n",170 "dtypes: object(6)\n",171 "memory usage: 41.9+ KB\n"172 ]173 }174 ],175 "source": [176 "df.info()"177 ]178 },179 {180 "cell_type": "markdown",181 "metadata": {},182 "source": [183 "### **Quality Issues** \n",184 "- Year has many non numeric values.\n",185 "- Year is in object data type.\n",186 "- Price has 35 Ask for Price values. \n",187 "- Price is in object data type.\n",188 "- Price has one outlier. \n",189 "- Kilometers Driven is in object data type and has one Petrol Entry.\n",190 "- Fuel Type has nan values."191 ]192 },193 {194 "cell_type": "markdown",195 "metadata": {},196 "source": [197 "### **Cleaning**"198 ]199 },200 {201 "cell_type": "code",202 "execution_count": 35,203 "metadata": {},204 "outputs": [],205 "source": [206 "df = df[df['year'].str.isnumeric()]"207 ]208 },209 {210 "cell_type": "code",211 "execution_count": 36,212 "metadata": {},213 "outputs": [],214 "source": [215 "df['year'] = df['year'].astype(int)"216 ]217 },218 {219 "cell_type": "code",220 "execution_count": 37,221 "metadata": {},222 "outputs": [],223 "source": [224 "df = df[df['Price'] != 'Ask For Price']"225 ]226 },227 {228 "cell_type": "code",229 "execution_count": 38,230 "metadata": {},231 "outputs": [],232 "source": [233 "df['Price'] = df[\"Price\"].str.replace(',','').astype(int)"234 ]235 },236 {237 "cell_type": "code",238 "execution_count": 39,239 "metadata": {},240 "outputs": [],241 "source": [242 "df = df[df['Price'] < 6e6]"243 ]244 },245 {246 "cell_type": "code",247 "execution_count": 40,248 "metadata": {},249 "outputs": [],250 "source": [251 "df = df[df['kms_driven'] != 'Petrol']"252 ]253 },254 {255 "cell_type": "code",256 "execution_count": 41,257 "metadata": {},258 "outputs": [],259 "source": [260 "df['kms_driven'] = df['kms_driven'].str.split(' ').str.get(0).str.replace(\",\",\"\").astype(int)"261 ]262 },263 {264 "cell_type": "code",265 "execution_count": 42,266 "metadata": {},267 "outputs": [],268 "source": [269 "df = df[~df['fuel_type'].isna()]"270 ]271 },272 {273 "cell_type": "code",274 "execution_count": 43,275 "metadata": {},276 "outputs": [],277 "source": [278 "df['name'] = df['name'].str.split(' ').str.slice(0,3).str.join(' ')"279 ]280 },281 {282 "cell_type": "code",283 "execution_count": 44,284 "metadata": {},285 "outputs": [286 {287 "data": {288 "text/html": [289 "<div>\n",290 "<style scoped>\n",291 " .dataframe tbody tr th:only-of-type {\n",292 " vertical-align: middle;\n",293 " }\n",294 "\n",295 " .dataframe tbody tr th {\n",296 " vertical-align: top;\n",297 " }\n",298 "\n",299 " .dataframe thead th {\n",300 " text-align: right;\n",301 " }\n",302 "</style>\n",303 "<table border=\"1\" class=\"dataframe\">\n",304 " <thead>\n",305 " <tr style=\"text-align: right;\">\n",306 " <th></th>\n",307 " <th>year</th>\n",308 " <th>Price</th>\n",309 " <th>kms_driven</th>\n",310 " </tr>\n",311 " </thead>\n",312 " <tbody>\n",313 " <tr>\n",314 " <th>count</th>\n",315 " <td>815.000000</td>\n",316 " <td>8.150000e+02</td>\n",317 " <td>815.000000</td>\n",318 " </tr>\n",319 " <tr>\n",320 " <th>mean</th>\n",321 " <td>2012.442945</td>\n",322 " <td>4.017933e+05</td>\n",323 " <td>46277.096933</td>\n",324 " </tr>\n",325 " <tr>\n",326 " <th>std</th>\n",327 " <td>4.005079</td>\n",328 " <td>3.815888e+05</td>\n",329 " <td>34318.459638</td>\n",330 " </tr>\n",331 " <tr>\n",332 " <th>min</th>\n",333 " <td>1995.000000</td>\n",334 " <td>3.000000e+04</td>\n",335 " <td>0.000000</td>\n",336 " </tr>\n",337 " <tr>\n",338 " <th>25%</th>\n",339 " <td>2010.000000</td>\n",340 " <td>1.750000e+05</td>\n",341 " <td>27000.000000</td>\n",342 " </tr>\n",343 " <tr>\n",344 " <th>50%</th>\n",345 " <td>2013.000000</td>\n",346 " <td>2.999990e+05</td>\n",347 " <td>41000.000000</td>\n",348 " </tr>\n",349 " <tr>\n",350 " <th>75%</th>\n",351 " <td>2015.000000</td>\n",352 " <td>4.900000e+05</td>\n",353 " <td>56879.000000</td>\n",354 " </tr>\n",355 " <tr>\n",356 " <th>max</th>\n",357 " <td>2019.000000</td>\n",358 " <td>3.100000e+06</td>\n",359 " <td>400000.000000</td>\n",360 " </tr>\n",361 " </tbody>\n",362 "</table>\n",363 "</div>"364 ],365 "text/plain": [366 " year Price kms_driven\n",367 "count 815.000000 8.150000e+02 815.000000\n",368 "mean 2012.442945 4.017933e+05 46277.096933\n",369 "std 4.005079 3.815888e+05 34318.459638\n",370 "min 1995.000000 3.000000e+04 0.000000\n",371 "25% 2010.000000 1.750000e+05 27000.000000\n",372 "50% 2013.000000 2.999990e+05 41000.000000\n",373 "75% 2015.000000 4.900000e+05 56879.000000\n",374 "max 2019.000000 3.100000e+06 400000.000000"375 ]376 },377 "execution_count": 44,378 "metadata": {},379 "output_type": "execute_result"380 }381 ],382 "source": [383 "df.describe()"384 ]385 },386 {387 "cell_type": "code",388 "execution_count": 45,389 "metadata": {},390 "outputs": [],391 "source": [392 "df = df.reset_index(drop=True)"393 ]394 },395 {396 "cell_type": "code",397 "execution_count": 46,398 "metadata": {},399 "outputs": [],400 "source": [401 "df.to_csv('clean_quikr_car.csv')"402 ]403 },404 {405 "cell_type": "code",406 "execution_count": 89,407 "metadata": {},408 "outputs": [409 {410 "data": {411 "text/plain": [412 "name\n",413 "Maruti Suzuki Swift 51\n",414 "Maruti Suzuki Alto 42\n",415 "Maruti Suzuki Wagon 28\n",416 "Maruti Suzuki Ertiga 16\n",417 "Hyundai Santro Xing 15\n",418 " ..\n",419 "Mercedes Benz A 1\n",420 "Tata Manza ELAN 1\n",421 "Volkswagen Polo Comfortline 1\n",422 "Nissan Sunny 1\n",423 "Tata Zest XM 1\n",424 "Name: count, Length: 254, dtype: int64"425 ]426 },427 "execution_count": 89,428 "metadata": {},429 "output_type": "execute_result"430 }431 ],432 "source": [433 "df['name'].value_counts()"434 ]435 },436 {437 "cell_type": "markdown",438 "metadata": {},439 "source": [440 "### **Model**"441 ]442 },443 {444 "cell_type": "code",445 "execution_count": 48,446 "metadata": {},447 "outputs": [],448 "source": [449 "X = df.drop(columns=['Price'])\n",450 "y = df['Price']"451 ]452 },453 {454 "cell_type": "code",455 "execution_count": 49,456 "metadata": {},457 "outputs": [],458 "source": [459 "from sklearn.model_selection import train_test_split\n",460 "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=124)"461 ]462 },463 {464 "cell_type": "code",465 "execution_count": 50,466 "metadata": {},467 "outputs": [468 {469 "data": {470 "text/html": [471 "<div>\n",472 "<style scoped>\n",473 " .dataframe tbody tr th:only-of-type {\n",474 " vertical-align: middle;\n",475 " }\n",476 "\n",477 " .dataframe tbody tr th {\n",478 " vertical-align: top;\n",479 " }\n",480 "\n",481 " .dataframe thead th {\n",482 " text-align: right;\n",483 " }\n",484 "</style>\n",485 "<table border=\"1\" class=\"dataframe\">\n",486 " <thead>\n",487 " <tr style=\"text-align: right;\">\n",488 " <th></th>\n",489 " <th>name</th>\n",490 " <th>company</th>\n",491 " <th>year</th>\n",492 " <th>kms_driven</th>\n",493 " <th>fuel_type</th>\n",494 " </tr>\n",495 " </thead>\n",496 " <tbody>\n",497 " <tr>\n",498 " <th>467</th>\n",499 " <td>Maruti Suzuki Vitara</td>\n",500 " <td>Maruti</td>\n",501 " <td>2017</td>\n",502 " <td>36000</td>\n",503 " <td>Diesel</td>\n",504 " </tr>\n",505 " <tr>\n",506 " <th>212</th>\n",507 " <td>Maruti Suzuki Alto</td>\n",508 " <td>Maruti</td>\n",509 " <td>2015</td>\n",510 " <td>5000</td>\n",511 " <td>Petrol</td>\n",512 " </tr>\n",513 " <tr>\n",514 " <th>685</th>\n",515 " <td>Hyundai Santro</td>\n",516 " <td>Hyundai</td>\n",517 " <td>2003</td>\n",518 " <td>51000</td>\n",519 " <td>Petrol</td>\n",520 " </tr>\n",521 " <tr>\n",522 " <th>538</th>\n",523 " <td>Maruti Suzuki Alto</td>\n",524 " <td>Maruti</td>\n",525 " <td>2019</td>\n",526 " <td>9800</td>\n",527 " <td>Petrol</td>\n",528 " </tr>\n",529 " <tr>\n",530 " <th>786</th>\n",531 " <td>Hyundai Eon</td>\n",532 " <td>Hyundai</td>\n",533 " <td>2018</td>\n",534 " <td>25000</td>\n",535 " <td>Petrol</td>\n",536 " </tr>\n",537 " <tr>\n",538 " <th>...</th>\n",539 " <td>...</td>\n",540 " <td>...</td>\n",541 " <td>...</td>\n",542 " <td>...</td>\n",543 " <td>...</td>\n",544 " </tr>\n",545 " <tr>\n",546 " <th>681</th>\n",547 " <td>Hyundai Santro AE</td>\n",548 " <td>Hyundai</td>\n",549 " <td>2011</td>\n",550 " <td>45000</td>\n",551 " <td>Petrol</td>\n",552 " </tr>\n",553 " <tr>\n",554 " <th>135</th>\n",555 " <td>Toyota Corolla Altis</td>\n",556 " <td>Toyota</td>\n",557 " <td>2012</td>\n",558 " <td>59000</td>\n",559 " <td>Petrol</td>\n",560 " </tr>\n",561 " <tr>\n",562 " <th>17</th>\n",563 " <td>Maruti Suzuki Alto</td>\n",564 " <td>Maruti</td>\n",565 " <td>2014</td>\n",566 " <td>35550</td>\n",567 " <td>Petrol</td>\n",568 " </tr>\n",569 " <tr>\n",570 " <th>668</th>\n",571 " <td>Maruti Suzuki Swift</td>\n",572 " <td>Maruti</td>\n",573 " <td>2014</td>\n",574 " <td>11523</td>\n",575 " <td>Petrol</td>\n",576 " </tr>\n",577 " <tr>\n",578 " <th>462</th>\n",579 " <td>Maruti Suzuki 800</td>\n",580 " <td>Maruti</td>\n",581 " <td>2001</td>\n",582 " <td>81876</td>\n",583 " <td>Petrol</td>\n",584 " </tr>\n",585 " </tbody>\n",586 "</table>\n",587 "<p>652 rows × 5 columns</p>\n",588 "</div>"589 ],590 "text/plain": [591 " name company year kms_driven fuel_type\n",592 "467 Maruti Suzuki Vitara Maruti 2017 36000 Diesel\n",593 "212 Maruti Suzuki Alto Maruti 2015 5000 Petrol\n",594 "685 Hyundai Santro Hyundai 2003 51000 Petrol\n",595 "538 Maruti Suzuki Alto Maruti 2019 9800 Petrol\n",596 "786 Hyundai Eon Hyundai 2018 25000 Petrol\n",597 ".. ... ... ... ... ...\n",598 "681 Hyundai Santro AE Hyundai 2011 45000 Petrol\n",599 "135 Toyota Corolla Altis Toyota 2012 59000 Petrol\n",600 "17 Maruti Suzuki Alto Maruti 2014 35550 Petrol\n",601 "668 Maruti Suzuki Swift Maruti 2014 11523 Petrol\n",602 "462 Maruti Suzuki 800 Maruti 2001 81876 Petrol\n",603 "\n",604 "[652 rows x 5 columns]"605 ]606 },607 "execution_count": 50,608 "metadata": {},609 "output_type": "execute_result"610 }611 ],612 "source": [613 "X_train"614 ]615 },616 {617 "cell_type": "code",618 "execution_count": 72,619 "metadata": {},620 "outputs": [],621 "source": [622 "from sklearn.linear_model import LinearRegression \n",623 "from sklearn.metrics import r2_score\n",624 "from sklearn.preprocessing import OneHotEncoder, OrdinalEncoder\n",625 "from sklearn.compose import make_column_transformer\n",626 "from sklearn.pipeline import make_pipeline"627 ]628 },629 {630 "cell_type": "code",631 "execution_count": 84,632 "metadata": {},633 "outputs": [634 {635 "data": {636 "text/plain": [637 "0.834097589979544"638 ]639 },640 "execution_count": 84,641 "metadata": {},642 "output_type": "execute_result"643 }644 ],645 "source": [646 "ohe = OneHotEncoder()\n",647 "ohe.fit(X[['name', 'company', 'fuel_type']])\n",648 "\n",649 "transformer = make_column_transformer(\n",650 " (OneHotEncoder(categories=ohe.categories_), ['name', 'company', 'fuel_type']),\n",651 " remainder='passthrough')\n",652 "\n",653 "lr = LinearRegression()\n",654 "\n",655 "pipe = make_pipeline(transformer, lr)\n",656 "\n",657 "pipe.fit(X_train, y_train)\n",658 "\n",659 "y_pred = pipe.predict(X_test)\n",660 "\n",661 "r2_score(y_pred, y_test)"662 ]663 },664 {665 "cell_type": "code",666 "execution_count": 86,667 "metadata": {},668 "outputs": [],669 "source": [670 "import pickle"671 ]672 },673 {674 "cell_type": "code",675 "execution_count": 87,676 "metadata": {},677 "outputs": [],678 "source": [679 "pickle.dump(pipe, open('pipe.pkl', 'wb'))"680 ]681 },682 {683 "cell_type": "code",684 "execution_count": 32,685 "metadata": {},686 "outputs": [],687 "source": [688 "pickle.dump(df, open('df.pkl', 'wb'))"689 ]690 },691 {692 "cell_type": "code",693 "execution_count": 88,694 "metadata": {},695 "outputs": [696 {697 "data": {698 "text/plain": [699 "array([438777.10574272])"700 ]701 },702 "execution_count": 88,703 "metadata": {},704 "output_type": "execute_result"705 }706 ],707 "source": [708 "pipe.predict(pd.DataFrame([['Maruti Suzuki Swift', 'Maruti', 2019, 100, 'Petrol']], columns=['name','company','year','kms_driven','fuel_type']))"709 ]710 }711 ],712 "metadata": {713 "kernelspec": {714 "display_name": "myenv",715 "language": "python",716 "name": "python3"717 },718 "language_info": {719 "codemirror_mode": {720 "name": "ipython",721 "version": 3722 },723 "file_extension": ".py",724 "mimetype": "text/x-python",725 "name": "python",726 "nbconvert_exporter": "python",727 "pygments_lexer": "ipython3",728 "version": "3.12.5"729 }730 },731 "nbformat": 4,732 "nbformat_minor": 2733}734