From 8989d1b1f4c11ba1687670f6725ac9d012d33202 Mon Sep 17 00:00:00 2001 From: Pedro Carneiro Marques Date: Fri, 22 Dec 2023 12:26:31 +0000 Subject: [PATCH] [technical_challenge_DA] - Pedro Marques --- data/tech_challenge.ipynb | 523 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 523 insertions(+) create mode 100644 data/tech_challenge.ipynb diff --git a/data/tech_challenge.ipynb b/data/tech_challenge.ipynb new file mode 100644 index 0000000..06f714c --- /dev/null +++ b/data/tech_challenge.ipynb @@ -0,0 +1,523 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Import Libraries" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Load the dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 388 entries, 0 to 387\n", + "Data columns (total 12 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 distance 388 non-null float64\n", + " 1 consume 388 non-null float64\n", + " 2 speed 388 non-null int64 \n", + " 3 temp_inside 376 non-null float64\n", + " 4 temp_outside 388 non-null int64 \n", + " 5 specials 93 non-null object \n", + " 6 gas_type 388 non-null object \n", + " 7 AC 388 non-null int64 \n", + " 8 rain 388 non-null int64 \n", + " 9 sun 388 non-null int64 \n", + " 10 refill liters 13 non-null float64\n", + " 11 refill gas 13 non-null object \n", + "dtypes: float64(4), int64(5), object(3)\n", + "memory usage: 36.5+ KB\n", + "None\n" + ] + } + ], + "source": [ + "gas = pd.read_csv(\"measurements.csv\", decimal=',')\n", + "print(gas.info())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Data Management" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " distance consume speed temp_inside temp_outside \\\n", + "count 388.000000 388.000000 388.000000 388.000000 388.000000 \n", + "mean 19.652835 4.912371 41.927835 21.929521 11.358247 \n", + "std 22.667837 1.033172 13.598524 0.994666 6.991542 \n", + "min 1.300000 3.300000 14.000000 19.000000 -5.000000 \n", + "25% 11.800000 4.300000 32.750000 21.500000 7.000000 \n", + "50% 14.600000 4.700000 40.500000 22.000000 10.000000 \n", + "75% 19.000000 5.300000 50.000000 22.500000 16.000000 \n", + "max 216.100000 12.200000 90.000000 25.500000 31.000000 \n", + "\n", + " AC rain sun \n", + "count 388.000000 388.000000 388.000000 \n", + "mean 0.077320 0.123711 0.082474 \n", + "std 0.267443 0.329677 0.275441 \n", + "min 0.000000 0.000000 0.000000 \n", + "25% 0.000000 0.000000 0.000000 \n", + "50% 0.000000 0.000000 0.000000 \n", + "75% 0.000000 0.000000 0.000000 \n", + "max 1.000000 1.000000 1.000000 \n", + " distance consume speed temp_inside \\\n", + "distance 1.000000 -0.128967 0.562299 0.075178 \n", + "consume -0.128967 1.000000 -0.227866 -0.160623 \n", + "speed 0.562299 -0.227866 1.000000 0.059293 \n", + "temp_inside 0.075178 -0.160623 0.059293 1.000000 \n", + "temp_outside 0.088175 -0.320811 0.015411 0.359500 \n", + "AC -0.025738 0.096591 -0.035408 0.297376 \n", + "rain -0.019791 0.248118 0.009489 -0.036937 \n", + "sun 0.081120 -0.170667 0.081618 0.242248 \n", + "gas_type_E10 0.053411 0.015327 0.097360 -0.010198 \n", + "gas_type_SP98 -0.053411 -0.015327 -0.097360 0.010198 \n", + "specials_AC -0.043842 -0.027819 -0.076235 0.187621 \n", + "specials_AC Sun 0.004148 -0.005536 0.018985 0.157121 \n", + "specials_AC rain -0.042487 0.086100 -0.034482 -0.066630 \n", + "specials_AC snow -0.016510 0.107772 0.037700 0.029192 \n", + "specials_AC sun 0.166715 -0.005536 0.172445 0.054778 \n", + "specials_ac 0.088429 -0.031621 -0.012584 0.247645 \n", + "specials_ac rain -0.007978 0.009243 -0.044645 0.157121 \n", + "specials_half rain half sun 0.053996 -0.010462 0.007756 0.003606 \n", + "specials_rain 0.010723 0.157122 0.038156 -0.068993 \n", + "specials_snow -0.029320 0.021755 0.015635 0.095124 \n", + "specials_sun 0.055737 -0.168194 0.033523 0.151238 \n", + "specials_sun ac -0.006055 -0.040982 0.048134 0.198796 \n", + "\n", + " temp_outside AC rain sun \\\n", + "distance 0.088175 -0.025738 -0.019791 0.081120 \n", + "consume -0.320811 0.096591 0.248118 -0.170667 \n", + "speed 0.015411 -0.035408 0.009489 0.081618 \n", + "temp_inside 0.359500 0.297376 -0.036937 0.242248 \n", + "temp_outside 1.000000 0.167562 -0.186315 0.346903 \n", + "AC 0.167562 1.000000 0.242915 0.088598 \n", + "rain -0.186315 0.242915 1.000000 -0.112650 \n", + "sun 0.346903 0.088598 -0.112650 1.000000 \n", + "gas_type_E10 -0.148705 -0.105285 -0.060328 -0.022761 \n", + "gas_type_SP98 0.148705 0.105285 0.060328 0.022761 \n", + "specials_AC 0.140150 0.432937 -0.047090 -0.037575 \n", + "specials_AC Sun 0.113872 0.175600 -0.019100 0.169549 \n", + "specials_AC rain -0.074114 0.532332 0.410129 -0.046201 \n", + "specials_AC snow -0.082688 0.175600 0.135289 -0.015240 \n", + "specials_AC sun 0.048352 0.175600 -0.019100 0.169549 \n", + "specials_ac 0.187366 0.433323 -0.054517 -0.043501 \n", + "specials_ac rain 0.033792 0.175600 0.135289 -0.015240 \n", + "specials_half rain half sun -0.017168 -0.014715 0.135289 -0.015240 \n", + "specials_rain -0.122726 -0.086790 0.797937 -0.089888 \n", + "specials_snow -0.139378 -0.025553 0.234936 -0.026466 \n", + "specials_sun 0.300752 -0.079168 -0.102757 0.912175 \n", + "specials_sun ac 0.121892 0.304938 -0.033167 0.294429 \n", + "\n", + " gas_type_E10 gas_type_SP98 ... \\\n", + "distance 0.053411 -0.053411 ... \n", + "consume 0.015327 -0.015327 ... \n", + "speed 0.097360 -0.097360 ... \n", + "temp_inside -0.010198 0.010198 ... \n", + "temp_outside -0.148705 0.148705 ... \n", + "AC -0.105285 0.105285 ... \n", + "rain -0.060328 0.060328 ... \n", + "sun -0.022761 0.022761 ... \n", + "gas_type_E10 1.000000 -1.000000 ... \n", + "gas_type_SP98 -1.000000 1.000000 ... \n", + "specials_AC -0.062555 0.062555 ... \n", + "specials_AC Sun 0.060681 -0.060681 ... \n", + "specials_AC rain -0.059525 0.059525 ... \n", + "specials_AC snow 0.060681 -0.060681 ... \n", + "specials_AC sun 0.060681 -0.060681 ... \n", + "specials_ac -0.084703 0.084703 ... \n", + "specials_ac rain 0.060681 -0.060681 ... \n", + "specials_half rain half sun -0.042583 0.042583 ... \n", + "specials_rain -0.022761 0.022761 ... \n", + "specials_snow -0.073947 0.073947 ... \n", + "specials_sun -0.023334 0.023334 ... \n", + "specials_sun ac -0.073947 0.073947 ... \n", + "\n", + " specials_AC rain specials_AC snow \\\n", + "distance -0.042487 -0.016510 \n", + "consume 0.086100 0.107772 \n", + "speed -0.034482 0.037700 \n", + "temp_inside -0.066630 0.029192 \n", + "temp_outside -0.074114 -0.082688 \n", + "AC 0.532332 0.175600 \n", + "rain 0.410129 0.135289 \n", + "sun -0.046201 -0.015240 \n", + "gas_type_E10 -0.059525 0.060681 \n", + "gas_type_SP98 0.059525 -0.060681 \n", + "specials_AC -0.019313 -0.006371 \n", + "specials_AC Sun -0.007833 -0.002584 \n", + "specials_AC rain 1.000000 -0.007833 \n", + "specials_AC snow -0.007833 1.000000 \n", + "specials_AC sun -0.007833 -0.002584 \n", + "specials_ac -0.022359 -0.007376 \n", + "specials_ac rain -0.007833 -0.002584 \n", + "specials_half rain half sun -0.007833 -0.002584 \n", + "specials_rain -0.046201 -0.015240 \n", + "specials_snow -0.013603 -0.004487 \n", + "specials_sun -0.042143 -0.013902 \n", + "specials_sun ac -0.013603 -0.004487 \n", + "\n", + " specials_AC sun specials_ac specials_ac rain \\\n", + "distance 0.166715 0.088429 -0.007978 \n", + "consume -0.005536 -0.031621 0.009243 \n", + "speed 0.172445 -0.012584 -0.044645 \n", + "temp_inside 0.054778 0.247645 0.157121 \n", + "temp_outside 0.048352 0.187366 0.033792 \n", + "AC 0.175600 0.433323 0.175600 \n", + "rain -0.019100 -0.054517 0.135289 \n", + "sun 0.169549 -0.043501 -0.015240 \n", + "gas_type_E10 0.060681 -0.084703 0.060681 \n", + "gas_type_SP98 -0.060681 0.084703 -0.060681 \n", + "specials_AC -0.006371 -0.018184 -0.006371 \n", + "specials_AC Sun -0.002584 -0.007376 -0.002584 \n", + "specials_AC rain -0.007833 -0.022359 -0.007833 \n", + "specials_AC snow -0.002584 -0.007376 -0.002584 \n", + "specials_AC sun 1.000000 -0.007376 -0.002584 \n", + "specials_ac -0.007376 1.000000 -0.007376 \n", + "specials_ac rain -0.002584 -0.007376 1.000000 \n", + "specials_half rain half sun -0.002584 -0.007376 -0.002584 \n", + "specials_rain -0.015240 -0.043501 -0.015240 \n", + "specials_snow -0.004487 -0.012808 -0.004487 \n", + "specials_sun -0.013902 -0.039681 -0.013902 \n", + "specials_sun ac -0.004487 -0.012808 -0.004487 \n", + "\n", + " specials_half rain half sun specials_rain \\\n", + "distance 0.053996 0.010723 \n", + "consume -0.010462 0.157122 \n", + "speed 0.007756 0.038156 \n", + "temp_inside 0.003606 -0.068993 \n", + "temp_outside -0.017168 -0.122726 \n", + "AC -0.014715 -0.086790 \n", + "rain 0.135289 0.797937 \n", + "sun -0.015240 -0.089888 \n", + "gas_type_E10 -0.042583 -0.022761 \n", + "gas_type_SP98 0.042583 0.022761 \n", + "specials_AC -0.006371 -0.037575 \n", + "specials_AC Sun -0.002584 -0.015240 \n", + "specials_AC rain -0.007833 -0.046201 \n", + "specials_AC snow -0.002584 -0.015240 \n", + "specials_AC sun -0.002584 -0.015240 \n", + "specials_ac -0.007376 -0.043501 \n", + "specials_ac rain -0.002584 -0.015240 \n", + "specials_half rain half sun 1.000000 -0.015240 \n", + "specials_rain -0.015240 1.000000 \n", + "specials_snow -0.004487 -0.026466 \n", + "specials_sun -0.013902 -0.081993 \n", + "specials_sun ac -0.004487 -0.026466 \n", + "\n", + " specials_snow specials_sun specials_sun ac \n", + "distance -0.029320 0.055737 -0.006055 \n", + "consume 0.021755 -0.168194 -0.040982 \n", + "speed 0.015635 0.033523 0.048134 \n", + "temp_inside 0.095124 0.151238 0.198796 \n", + "temp_outside -0.139378 0.300752 0.121892 \n", + "AC -0.025553 -0.079168 0.304938 \n", + "rain 0.234936 -0.102757 -0.033167 \n", + "sun -0.026466 0.912175 0.294429 \n", + "gas_type_E10 -0.073947 -0.023334 -0.073947 \n", + "gas_type_SP98 0.073947 0.023334 0.073947 \n", + "specials_AC -0.011063 -0.034275 -0.011063 \n", + "specials_AC Sun -0.004487 -0.013902 -0.004487 \n", + "specials_AC rain -0.013603 -0.042143 -0.013603 \n", + "specials_AC snow -0.004487 -0.013902 -0.004487 \n", + "specials_AC sun -0.004487 -0.013902 -0.004487 \n", + "specials_ac -0.012808 -0.039681 -0.012808 \n", + "specials_ac rain -0.004487 -0.013902 -0.004487 \n", + "specials_half rain half sun -0.004487 -0.013902 -0.004487 \n", + "specials_rain -0.026466 -0.081993 -0.026466 \n", + "specials_snow 1.000000 -0.024141 -0.007792 \n", + "specials_sun -0.024141 1.000000 -0.024141 \n", + "specials_sun ac -0.007792 -0.024141 1.000000 \n", + "\n", + "[22 rows x 22 columns]\n" + ] + } + ], + "source": [ + "# Handling missing values\n", + "# Drop 'refill liters' and 'refill gas' due to high null cound\n", + "gas.drop(['refill liters', 'refill gas'], axis=1, inplace=True)\n", + "\n", + "# Impute missing values in 'temp_inside' with the mean\n", + "gas['temp_inside'].fillna(gas['temp_inside'].mean(), inplace=True)\n", + "\n", + "# Encoding categorical variables\n", + "data = pd.get_dummies(gas, columns=['gas_type', 'specials'])\n", + "\n", + "# Summary statistics\n", + "print(data.describe())\n", + "\n", + "# Correlation matrix\n", + "print(data.corr())\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Data Visualization" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/slevin/anaconda3/lib/python3.11/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n", + " if pd.api.types.is_categorical_dtype(vector):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n", + " if pd.api.types.is_categorical_dtype(vector):\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/slevin/anaconda3/lib/python3.11/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n", + " if pd.api.types.is_categorical_dtype(vector):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/seaborn/_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n", + " with pd.option_context('mode.use_inf_as_na', True):\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/slevin/anaconda3/lib/python3.11/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n", + " if pd.api.types.is_categorical_dtype(vector):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n", + " if pd.api.types.is_categorical_dtype(vector):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n", + " if pd.api.types.is_categorical_dtype(vector):\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Scatter plot: Consumption vs. Speed\n", + "plt.figure(figsize=(8, 6))\n", + "sns.scatterplot(x='speed', y='consume', data=data)\n", + "plt.title('Fuel Consumption vs. Speed')\n", + "plt.xlabel('Speed')\n", + "plt.ylabel('Fuel Consumption')\n", + "plt.show()\n", + "\n", + "# Histogram: Distribution of Consumption\n", + "plt.figure(figsize=(8, 6))\n", + "sns.histplot(data['consume'], bins=20, kde=True)\n", + "plt.title('Distribution of Fuel Consumption')\n", + "plt.xlabel('Fuel Consumption')\n", + "plt.ylabel('Frequency')\n", + "plt.show()\n", + "\n", + "# Box plot: Consumption vs. Gas Type\n", + "plt.figure(figsize=(8, 6))\n", + "sns.boxplot(x='gas_type_E10', y='consume', data=data)\n", + "plt.title('Fuel Consumption by Gas Type')\n", + "plt.xlabel('Gas Type (E10 = 0, SP98 = 1)')\n", + "plt.ylabel('Fuel Consumption')\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Machine Learning Model" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Mean Squared Error: 0.8216359860810033\n", + "R^2 Score: 0.0944224528189378\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:767: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if not hasattr(array, \"sparse\") and array.dtypes.apply(is_sparse).any():\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:605: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:614: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype) or not is_extension_array_dtype(pd_dtype):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:605: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:614: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype) or not is_extension_array_dtype(pd_dtype):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:767: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if not hasattr(array, \"sparse\") and array.dtypes.apply(is_sparse).any():\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:605: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:614: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype) or not is_extension_array_dtype(pd_dtype):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:605: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:614: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype) or not is_extension_array_dtype(pd_dtype):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:605: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype):\n", + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/utils/validation.py:614: FutureWarning: is_sparse is deprecated and will be removed in a future version. Check `isinstance(dtype, pd.SparseDtype)` instead.\n", + " if is_sparse(pd_dtype) or not is_extension_array_dtype(pd_dtype):\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", + "\n", + "# Feature selection and split\n", + "features = ['distance', 'speed', 'temp_inside', 'temp_outside', 'AC', 'rain', 'sun', 'gas_type_E10', 'gas_type_SP98']\n", + "X = data[features]\n", + "y = data['consume']\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", + "\n", + "# Linear Regression model\n", + "model = LinearRegression()\n", + "model.fit(X_train, y_train)\n", + "\n", + "# Predictions\n", + "y_pred = model.predict(X_test)\n", + "\n", + "# Model evaluation\n", + "mse = mean_squared_error(y_test, y_pred)\n", + "r2 = r2_score(y_test, y_pred)\n", + "print(f\"Mean Squared Error: {mse}\")\n", + "print(f\"R^2 Score: {r2}\")" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Predicted fuel consumption: 4.26987685029525\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/slevin/anaconda3/lib/python3.11/site-packages/sklearn/base.py:439: UserWarning: X does not have valid feature names, but LinearRegression was fitted with feature names\n", + " warnings.warn(\n" + ] + } + ], + "source": [ + "new_trip = [[20, 60, 22, 15, 0, 0, 1, 1, 0]] # Example data for distance, speed, temp_inside, temp_outside, AC, rain, sun, gas_type_E10, gas_type_SP98\n", + "predicted_consumption = model.predict(new_trip)\n", + "print(f\"Predicted fuel consumption: {predicted_consumption[0]}\")" + ] + }, + { + "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": "3.11.0" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}