{ "cells": [ { "cell_type": "markdown", "id": "f35a75ba", "metadata": {}, "source": [ "" ] }, { "cell_type": "markdown", "id": "7a8ad074", "metadata": {}, "source": [ "# Regresión y Clasificación con Redes Neuronales (MLP)" ] }, { "cell_type": "code", "execution_count": 680, "id": "23d79b8b", "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.metrics import classification_report,confusion_matrix,ConfusionMatrixDisplay\n", "from sklearn import model_selection as ms\n", "from sklearn.metrics import r2_score \n", "from sklearn.metrics import mean_squared_error\n", "from sklearn import metrics as m\n", "from sklearn.neural_network import MLPRegressor\n", "from sklearn.neural_network import MLPClassifier\n", "from sklearn.preprocessing import StandardScaler\n" ] }, { "cell_type": "markdown", "id": "7058d734", "metadata": {}, "source": [ "# 1. Regresión" ] }, { "cell_type": "markdown", "id": "59941cf1", "metadata": {}, "source": [ "### Información del dataset" ] }, { "cell_type": "markdown", "id": "c24fba07", "metadata": {}, "source": [ " ### Bike Rents for the Day Dataset\n", " \n", " https://www.kaggle.com/datasets/ayessa/bike-sharing-dataset-regression\n", " \n", " \n", " " ] }, { "cell_type": "markdown", "id": "6a83e681", "metadata": {}, "source": [ "Este conjunto de datos contiene el recuento diario y por hora de bicicletas de alquiler entre los años 2011 y 2012 en el sistema de bicicletas compartidas de Capital con la información meteorológica y estacional correspondiente.\n", "\n", "\n", "Los sistemas de bicicletas compartidas son una nueva generación de alquileres de bicicletas tradicionales en los que todo el proceso, desde la afiliación, el alquiler y la devolución, se ha vuelto automático. A través de estos sistemas, el usuario puede alquilar fácilmente una bicicleta desde una posición particular y regresar en otra posición. Actualmente, hay alrededor de 500 programas de bicicletas compartidas en todo el mundo que se componen de más de 500 mil bicicletas. Hoy en día, existe un gran interés en estos sistemas debido a su importante papel en cuestiones de tráfico, medio ambiente y salud.\n", "\n", "Aparte de las interesantes aplicaciones del mundo real de los sistemas de bicicletas compartidas, las características de los datos generados por estos sistemas los hacen atractivos para la investigación. A diferencia de otros servicios de transporte como el autobús o el metro, en estos sistemas se registra explícitamente la duración del viaje, la posición de salida y de llegada. Esta función convierte el sistema de bicicletas compartidas en una red de sensores virtual que se puede utilizar para detectar la movilidad en la ciudad. Por lo tanto, se espera que la mayoría de los eventos importantes en la ciudad puedan detectarse a través del monitoreo de estos datos.\n", "\n", "\n", "Las columnas (Características) son:\n", "\n", "\n", "- instant: índice de registros\n", "- dteday : fecha\n", "- season : estación (1:invierno, 2:primavera, 3:verano, 4:otoño)\n", "- yr : año (0: 2011, 1:2012)\n", "- mnth : mes ( 1 a 12)\n", "- hr : hora (0 a 23)\n", "- holiday : si el día es festivo o no (extraído de [Web Link])\n", "- weekday : día de la semana\n", "- workingday : si el día no es ni fin de semana ni festivo es 1, en caso contrario es 0.\n", "+ weathersit : 1: Despejado, Pocas nubes, Parcialmente nublado, Parcialmente nublado, 2: Niebla + Nublado, Niebla + Nubes dispersas, Niebla + Pocas nubes, Niebla, 3: Nieve ligera, Lluvia ligera + Tormenta + Nubes dispersas, Lluvia ligera + Nubes dispersas, 4: Lluvia intensa + Paletas de hielo + Tormenta eléctrica + Nieve, Nieve + Niebla.\n", "- temp : Temperatura normalizada en Celsius. Los valores se obtienen mediante (t-t_min)/(t_max-t_min), t_min=-8, t_max=+39 (sólo en escala horaria)\n", "- atemp: Temperatura de sensación normalizada en Celsius. Los valores se obtienen mediante (t-t_min)/(t_max-t_min), t_min=-16, t_max=+50 (sólo en escala horaria)\n", "- hum: Humedad normalizada. Los valores se dividen entre 100 (máximo)\n", "- windspeed: Velocidad del viento normalizada. Los valores se dividen entre 67 (máx.)\n", "- casual: recuento de usuarios ocasionales\n", "- registered: recuento de usuarios registrados\n", "- cnt: recuento del total de bicicletas de alquiler, incluyendo las casuales y las registradas\n" ] }, { "cell_type": "markdown", "id": "6fd4e790", "metadata": {}, "source": [ "### Tarea\n", "\n", "Predecir el número de bicicletas alquiladas en un día, en función de la temperatura, el día y más." ] }, { "cell_type": "markdown", "id": "e132db38", "metadata": {}, "source": [ "### 1 Análisis exploratorio de los datos" ] }, { "cell_type": "markdown", "id": "4a818f11", "metadata": {}, "source": [ "1. Imprima el número de registros del dataset\n", "2. Imprima el número de variables del dataset\n", "3. Imprima el nombre de las columnas del dataset\n", "4. Imprima el **head** del dataset\n", "5. Imprima el **tail** del dataset\n", "6. Imprima **info** basica del dataset\n", "7. Imprima un **describe** del dataset" ] }, { "cell_type": "code", "execution_count": 78, "id": "0f981d48", "metadata": {}, "outputs": [], "source": [ "data = pd.read_csv(\"./resources/day.csv\")" ] }, { "cell_type": "code", "execution_count": 79, "id": "290328fb", "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", "
instantdtedayseasonyrmnthholidayweekdayworkingdayweathersittempatemphumwindspeedcasualregisteredcnt
012011-01-0110106020.3441670.3636250.8058330.160446331654985
122011-01-0210100020.3634780.3537390.6960870.248539131670801
232011-01-0310101110.1963640.1894050.4372730.24830912012291349
342011-01-0410102110.2000000.2121220.5904350.16029610814541562
452011-01-0510103110.2269570.2292700.4369570.1869008215181600
...................................................
7267272012-12-27111204120.2541670.2266420.6529170.35013324718672114
7277282012-12-28111205120.2533330.2550460.5900000.15547164424513095
7287292012-12-29111206020.2533330.2424000.7529170.12438315911821341
7297302012-12-30111200010.2558330.2317000.4833330.35075436414321796
7307312012-12-31111201120.2158330.2234870.5775000.15484643922902729
\n", "

731 rows × 16 columns

\n", "
" ], "text/plain": [ " instant dteday season yr mnth holiday weekday workingday \\\n", "0 1 2011-01-01 1 0 1 0 6 0 \n", "1 2 2011-01-02 1 0 1 0 0 0 \n", "2 3 2011-01-03 1 0 1 0 1 1 \n", "3 4 2011-01-04 1 0 1 0 2 1 \n", "4 5 2011-01-05 1 0 1 0 3 1 \n", ".. ... ... ... .. ... ... ... ... \n", "726 727 2012-12-27 1 1 12 0 4 1 \n", "727 728 2012-12-28 1 1 12 0 5 1 \n", "728 729 2012-12-29 1 1 12 0 6 0 \n", "729 730 2012-12-30 1 1 12 0 0 0 \n", "730 731 2012-12-31 1 1 12 0 1 1 \n", "\n", " weathersit temp atemp hum windspeed casual registered \\\n", "0 2 0.344167 0.363625 0.805833 0.160446 331 654 \n", "1 2 0.363478 0.353739 0.696087 0.248539 131 670 \n", "2 1 0.196364 0.189405 0.437273 0.248309 120 1229 \n", "3 1 0.200000 0.212122 0.590435 0.160296 108 1454 \n", "4 1 0.226957 0.229270 0.436957 0.186900 82 1518 \n", ".. ... ... ... ... ... ... ... \n", "726 2 0.254167 0.226642 0.652917 0.350133 247 1867 \n", "727 2 0.253333 0.255046 0.590000 0.155471 644 2451 \n", "728 2 0.253333 0.242400 0.752917 0.124383 159 1182 \n", "729 1 0.255833 0.231700 0.483333 0.350754 364 1432 \n", "730 2 0.215833 0.223487 0.577500 0.154846 439 2290 \n", "\n", " cnt \n", "0 985 \n", "1 801 \n", "2 1349 \n", "3 1562 \n", "4 1600 \n", ".. ... \n", "726 2114 \n", "727 3095 \n", "728 1341 \n", "729 1796 \n", "730 2729 \n", "\n", "[731 rows x 16 columns]" ] }, "execution_count": 79, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data" ] }, { "cell_type": "code", "execution_count": 80, "id": "e69d3aa7", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Número de registros 731\n" ] } ], "source": [ "print(\"Número de registros\",)" ] }, { "cell_type": "code", "execution_count": 81, "id": "8156e300", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Número de variables 16\n" ] } ], "source": [ "print(\"Número de variables\"," ] }, { "cell_type": "code", "execution_count": 82, "id": "eee05f8a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Index(['instant', 'dteday', 'season', 'yr', 'mnth', 'holiday', 'weekday',\n", " 'workingday', 'weathersit', 'temp', 'atemp', 'hum', 'windspeed',\n", " 'casual', 'registered', 'cnt'],\n", " dtype='object')\n" ] } ], "source": [] }, { "cell_type": "code", "execution_count": 83, "id": "443efe52", "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", "
instantdtedayseasonyrmnthholidayweekdayworkingdayweathersittempatemphumwindspeedcasualregisteredcnt
012011-01-0110106020.3441670.3636250.8058330.160446331654985
122011-01-0210100020.3634780.3537390.6960870.248539131670801
232011-01-0310101110.1963640.1894050.4372730.24830912012291349
342011-01-0410102110.2000000.2121220.5904350.16029610814541562
452011-01-0510103110.2269570.2292700.4369570.1869008215181600
\n", "
" ], "text/plain": [ " instant dteday season yr mnth holiday weekday workingday \\\n", "0 1 2011-01-01 1 0 1 0 6 0 \n", "1 2 2011-01-02 1 0 1 0 0 0 \n", "2 3 2011-01-03 1 0 1 0 1 1 \n", "3 4 2011-01-04 1 0 1 0 2 1 \n", "4 5 2011-01-05 1 0 1 0 3 1 \n", "\n", " weathersit temp atemp hum windspeed casual registered \\\n", "0 2 0.344167 0.363625 0.805833 0.160446 331 654 \n", "1 2 0.363478 0.353739 0.696087 0.248539 131 670 \n", "2 1 0.196364 0.189405 0.437273 0.248309 120 1229 \n", "3 1 0.200000 0.212122 0.590435 0.160296 108 1454 \n", "4 1 0.226957 0.229270 0.436957 0.186900 82 1518 \n", "\n", " cnt \n", "0 985 \n", "1 801 \n", "2 1349 \n", "3 1562 \n", "4 1600 " ] }, "execution_count": 83, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "code", "execution_count": 84, "id": "afc33abf", "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", "
instantdtedayseasonyrmnthholidayweekdayworkingdayweathersittempatemphumwindspeedcasualregisteredcnt
7267272012-12-27111204120.2541670.2266420.6529170.35013324718672114
7277282012-12-28111205120.2533330.2550460.5900000.15547164424513095
7287292012-12-29111206020.2533330.2424000.7529170.12438315911821341
7297302012-12-30111200010.2558330.2317000.4833330.35075436414321796
7307312012-12-31111201120.2158330.2234870.5775000.15484643922902729
\n", "
" ], "text/plain": [ " instant dteday season yr mnth holiday weekday workingday \\\n", "726 727 2012-12-27 1 1 12 0 4 1 \n", "727 728 2012-12-28 1 1 12 0 5 1 \n", "728 729 2012-12-29 1 1 12 0 6 0 \n", "729 730 2012-12-30 1 1 12 0 0 0 \n", "730 731 2012-12-31 1 1 12 0 1 1 \n", "\n", " weathersit temp atemp hum windspeed casual registered \\\n", "726 2 0.254167 0.226642 0.652917 0.350133 247 1867 \n", "727 2 0.253333 0.255046 0.590000 0.155471 644 2451 \n", "728 2 0.253333 0.242400 0.752917 0.124383 159 1182 \n", "729 1 0.255833 0.231700 0.483333 0.350754 364 1432 \n", "730 2 0.215833 0.223487 0.577500 0.154846 439 2290 \n", "\n", " cnt \n", "726 2114 \n", "727 3095 \n", "728 1341 \n", "729 1796 \n", "730 2729 " ] }, "execution_count": 84, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "code", "execution_count": 85, "id": "81f353dd", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RangeIndex: 731 entries, 0 to 730\n", "Data columns (total 16 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 instant 731 non-null int64 \n", " 1 dteday 731 non-null object \n", " 2 season 731 non-null int64 \n", " 3 yr 731 non-null int64 \n", " 4 mnth 731 non-null int64 \n", " 5 holiday 731 non-null int64 \n", " 6 weekday 731 non-null int64 \n", " 7 workingday 731 non-null int64 \n", " 8 weathersit 731 non-null int64 \n", " 9 temp 731 non-null float64\n", " 10 atemp 731 non-null float64\n", " 11 hum 731 non-null float64\n", " 12 windspeed 731 non-null float64\n", " 13 casual 731 non-null int64 \n", " 14 registered 731 non-null int64 \n", " 15 cnt 731 non-null int64 \n", "dtypes: float64(4), int64(11), object(1)\n", "memory usage: 91.5+ KB\n" ] } ], "source": [] }, { "cell_type": "code", "execution_count": 86, "id": "c6266239", "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", "
instantseasonyrmnthholidayweekdayworkingdayweathersittempatemphumwindspeedcasualregisteredcnt
count731.000000731.000000731.000000731.000000731.000000731.000000731.000000731.000000731.000000731.000000731.000000731.000000731.000000731.000000731.000000
mean366.0000002.4965800.5006846.5198360.0287282.9972640.6839951.3953490.4953850.4743540.6278940.190486848.1764713656.1723674504.348837
std211.1658121.1108070.5003423.4519130.1671552.0047870.4652330.5448940.1830510.1629610.1424290.077498686.6224881560.2563771937.211452
min1.0000001.0000000.0000001.0000000.0000000.0000000.0000001.0000000.0591300.0790700.0000000.0223922.00000020.00000022.000000
25%183.5000002.0000000.0000004.0000000.0000001.0000000.0000001.0000000.3370830.3378420.5200000.134950315.5000002497.0000003152.000000
50%366.0000003.0000001.0000007.0000000.0000003.0000001.0000001.0000000.4983330.4867330.6266670.180975713.0000003662.0000004548.000000
75%548.5000003.0000001.00000010.0000000.0000005.0000001.0000002.0000000.6554170.6086020.7302090.2332141096.0000004776.5000005956.000000
max731.0000004.0000001.00000012.0000001.0000006.0000001.0000003.0000000.8616670.8408960.9725000.5074633410.0000006946.0000008714.000000
\n", "
" ], "text/plain": [ " instant season yr mnth holiday weekday \\\n", "count 731.000000 731.000000 731.000000 731.000000 731.000000 731.000000 \n", "mean 366.000000 2.496580 0.500684 6.519836 0.028728 2.997264 \n", "std 211.165812 1.110807 0.500342 3.451913 0.167155 2.004787 \n", "min 1.000000 1.000000 0.000000 1.000000 0.000000 0.000000 \n", "25% 183.500000 2.000000 0.000000 4.000000 0.000000 1.000000 \n", "50% 366.000000 3.000000 1.000000 7.000000 0.000000 3.000000 \n", "75% 548.500000 3.000000 1.000000 10.000000 0.000000 5.000000 \n", "max 731.000000 4.000000 1.000000 12.000000 1.000000 6.000000 \n", "\n", " workingday weathersit temp atemp hum windspeed \\\n", "count 731.000000 731.000000 731.000000 731.000000 731.000000 731.000000 \n", "mean 0.683995 1.395349 0.495385 0.474354 0.627894 0.190486 \n", "std 0.465233 0.544894 0.183051 0.162961 0.142429 0.077498 \n", "min 0.000000 1.000000 0.059130 0.079070 0.000000 0.022392 \n", "25% 0.000000 1.000000 0.337083 0.337842 0.520000 0.134950 \n", "50% 1.000000 1.000000 0.498333 0.486733 0.626667 0.180975 \n", "75% 1.000000 2.000000 0.655417 0.608602 0.730209 0.233214 \n", "max 1.000000 3.000000 0.861667 0.840896 0.972500 0.507463 \n", "\n", " casual registered cnt \n", "count 731.000000 731.000000 731.000000 \n", "mean 848.176471 3656.172367 4504.348837 \n", "std 686.622488 1560.256377 1937.211452 \n", "min 2.000000 20.000000 22.000000 \n", "25% 315.500000 2497.000000 3152.000000 \n", "50% 713.000000 3662.000000 4548.000000 \n", "75% 1096.000000 4776.500000 5956.000000 \n", "max 3410.000000 6946.000000 8714.000000 " ] }, "execution_count": 86, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "code", "execution_count": 87, "id": "5db5b000", "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", "
instantdtedayseasonyrmnthholidayweekdayworkingdayweathersittempatemphumwindspeedcasualregisteredcnt
012011-01-0110106020.3441670.3636250.8058330.160446331654985
122011-01-0210100020.3634780.3537390.6960870.248539131670801
232011-01-0310101110.1963640.1894050.4372730.24830912012291349
\n", "
" ], "text/plain": [ " instant dteday season yr mnth holiday weekday workingday \\\n", "0 1 2011-01-01 1 0 1 0 6 0 \n", "1 2 2011-01-02 1 0 1 0 0 0 \n", "2 3 2011-01-03 1 0 1 0 1 1 \n", "\n", " weathersit temp atemp hum windspeed casual registered \\\n", "0 2 0.344167 0.363625 0.805833 0.160446 331 654 \n", "1 2 0.363478 0.353739 0.696087 0.248539 131 670 \n", "2 1 0.196364 0.189405 0.437273 0.248309 120 1229 \n", "\n", " cnt \n", "0 985 \n", "1 801 \n", "2 1349 " ] }, "execution_count": 87, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "markdown", "id": "07523317", "metadata": {}, "source": [ "8. Agrupar la información por estación(season) y mes(mnth) y sumar el número de bicicletas prestadas." ] }, { "cell_type": "code", "execution_count": 88, "id": "58c718ce", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "44f8e26f", "metadata": {}, "source": [ "9. Usando la información obtenida en el punto anterior(8), realizar un diagrama de barra con la distribución de prestamos de bicicletas por mes " ] }, { "cell_type": "code", "execution_count": 89, "id": "90d12645", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 89, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [] }, { "cell_type": "markdown", "id": "f5c2b027", "metadata": {}, "source": [ "10. Realizar un diagrama de dispersión relacionando las variables atemp y cnt" ] }, { "cell_type": "code", "execution_count": 90, "id": "4d1a4f39", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Text(0, 0.5, 'cnt')" ] }, "execution_count": 90, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [] }, { "cell_type": "markdown", "id": "15f84cae", "metadata": {}, "source": [ "11. Eliminar la columna **instant** ya que no aporta información relevante (es un indice)" ] }, { "cell_type": "code", "execution_count": 91, "id": "8890b2e4", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "5d66f01d", "metadata": {}, "source": [ "12. Realizar un mapa de calor usando Coeficiente de correlación de Pearson, ejecute la siguiente celda de código para lograr esto " ] }, { "cell_type": "code", "execution_count": 93, "id": "9b64c706", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 93, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "f, ax = plt.subplots(figsize=(10, 8))\n", "\n", "mask = np.triu(np.ones_like(data.corr()))\n", "sns.heatmap(data.corr('pearson'),annot = True,mask=mask)" ] }, { "cell_type": "markdown", "id": "a3bf4fe0", "metadata": {}, "source": [ "### 2. Tratamiento de missing, reparación dataset y codificación de variables" ] }, { "cell_type": "markdown", "id": "0a1415ed", "metadata": {}, "source": [ "1. Defina un vector **Y** con el valor de la columna **cnt** del dataset" ] }, { "cell_type": "code", "execution_count": 95, "id": "80d012b1", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "a9787700", "metadata": {}, "source": [ "2. Elimine las columnas casual, registered y cnt del dataset ya que la columna casual + registered = cnt, registered es el número de usuarios registrados y casual son el número de usuarios casuales, ambas variables sumadas dan como resultado la columna cnt que es la variable que se busca predecir." ] }, { "cell_type": "code", "execution_count": 99, "id": "d2317c44", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "8af07aeb", "metadata": {}, "source": [ "3. Defina un vector X solo usando aquellas variables que tengan un coeffiente de correlación de Pearsonsuperior a 0.2\n", "\n", "Nota: Mirar mapa de calor" ] }, { "cell_type": "code", "execution_count": 103, "id": "6d0a188c", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "8854e34f", "metadata": {}, "source": [ "### 3. Determinar el conjunto de entrenamiento y el de validación." ] }, { "cell_type": "markdown", "id": "2ba31431", "metadata": {}, "source": [ "1. Crear un vector X el cual contiene las características \n", "2. Crear un vector y el cual contiene las clases\n", "3. Imprimir el vector X\n", "4. Imprimir el vector y\n", "\n", "5. Hacer división de los datos 80% train , 20% test \n", "\n", "Ayuda: usar la función train_test_split de sklearn https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html\n", "\n", "6. Imprimir las dimensiones del conjunto de train y test\n" ] }, { "cell_type": "code", "execution_count": 104, "id": "e683cb50", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 105, "id": "27ae1add", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Dimensiones vector de entrenamiento (584, 5)\n", "Dimensiones vector de prueba (147, 5)\n" ] } ], "source": [ "print(\"Dimensiones vector de entrenamiento\", )\n", "print(\"Dimensiones vector de prueba\", )" ] }, { "cell_type": "markdown", "id": "8bb7bb81", "metadata": {}, "source": [ "### 4. Entrenamiento del modelo" ] }, { "cell_type": "markdown", "id": "13929f8d", "metadata": {}, "source": [ "1. Crear un MLPRegressor model usando la librería sklearn https://scikit-learn.org/stable/modules/generated/sklearn.neural_network.MLPRegressor.html\n", "2. Entrenar el modelo\n", "3. Usar 6 capas ocultas de 1000,1000,100,50,10,5\n", "4. Utilizar un max_iter = 500\n", "5. Usar early_stopping = True\n", "\n", "Ayudas:\n", "\n", "- Usar la función fit\n", "- Solo usar el conjunto de entrenamiento (X_train, y_train)" ] }, { "cell_type": "code", "execution_count": 385, "id": "f6e81333", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "64d88e97", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 395, "id": "ca460767", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Número de iteraciones necesarias para entrenar el modelo 263\n" ] } ], "source": [ "print(\"Número de iteraciones necesarias para entrenar el modelo\",)" ] }, { "cell_type": "markdown", "id": "603c3418", "metadata": {}, "source": [ "### 5. Calcular las métricas de evaluación" ] }, { "cell_type": "markdown", "id": "ee27e6df", "metadata": {}, "source": [ "1. Usar la función predict() para crear el vector de predicciones\n", "\n", "Ayuda: Utilice el conjunto de test (X_test)" ] }, { "cell_type": "code", "execution_count": 389, "id": "1dd4d731", "metadata": {}, "outputs": [], "source": [ "y_predict = " ] }, { "cell_type": "markdown", "id": "60b91a8a", "metadata": {}, "source": [ "**Nota:** Ejecutar la siguiente celda la cual calcula diversas métricas para problemas de regresión. " ] }, { "cell_type": "code", "execution_count": 392, "id": "7c576904", "metadata": {}, "outputs": [], "source": [ "mae_test = m.mean_absolute_error(y_test, y_predict )\n", "mape_test = np.mean(np.abs((y_test - y_predict)/ y_test))\n", "MSE_test = mean_squared_error(y_test,y_predict)\n", "RMSE_test = mean_squared_error(y_test,y_predict,squared=False) \n", "R2_test = r2_score(y_test,y_predict)" ] }, { "cell_type": "code", "execution_count": 393, "id": "618b0213", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "MAE 729.0662072296108\n", "MAPE 0.2415938813321013\n", "MSE 917387.7243668467\n", "RMSE 957.803593836882\n", "R2 0.7596922000879675\n" ] } ], "source": [ "print(\"MAE\",mae_test)\n", "print(\"MAPE\",mape_test)\n", "print(\"MSE\",MSE_test)\n", "print(\"RMSE\",RMSE_test)\n", "print(\"R2\",R2_test)" ] }, { "cell_type": "markdown", "id": "79c46a3c", "metadata": {}, "source": [ "### 6. Conclusiones" ] }, { "cell_type": "markdown", "id": "016441ec", "metadata": {}, "source": [ "1. Describa brevemente los resultados obtenidos" ] }, { "cell_type": "markdown", "id": "1e21d313", "metadata": {}, "source": [ "El modelo presento un desempeño de 24% de Error Promedio Porcentual Medio (MAPE) y un R-cuadrado de 0.76 lo cual indica una correlación lineal positiva ya que tiende a 1 sin embargo el modelo presenta un desempeño regular se tendría que experimentar y jugar con el número de capas y neuronas por capas, etc." ] }, { "cell_type": "markdown", "id": "7fa68821", "metadata": {}, "source": [ "2. Realizar un gráfico de dispersión entre y_test y y_predict" ] }, { "cell_type": "code", "execution_count": 396, "id": "ed3c003b", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 396, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [] }, { "cell_type": "markdown", "id": "4c488be0", "metadata": {}, "source": [ "# 2. Clasificación" ] }, { "cell_type": "markdown", "id": "a89673df", "metadata": {}, "source": [ "### Información del dataset\n", "\n", "\n", "#### Mobile Price Classification\n", "\n", "https://www.kaggle.com/datasets/iabhishekofficial/mobile-price-classification\n", "\n", "\n", "Bob ha comenzado su propia empresa de telefonía móvil. Quiere dar una pelea dura a las grandes empresas como Apple, Samsung, etc. No sabe cómo estimar el precio de los móviles que fabrica su empresa. En este competitivo mercado de telefonía móvil no se puede simplemente asumir cosas. Para resolver este problema, recopila datos de ventas de teléfonos móviles de varias empresas. Bob quiere averiguar alguna relación entre las funciones de un teléfono móvil (p. ej., RAM, memoria interna, etc.) y su precio de venta. Pero no es tan bueno en Machine Learning. Así que necesita tu ayuda para resolver este problema. En este problema, no tiene que predecir el precio real, sino un rango de precios que indica qué tan alto es el precio.\n", "\n", "\n", "Price_range: Esta es la variable objetivo con valor de 0 (costo bajo), 1 (costo medio), 2 (costo alto) y 3 (costo muy alto).\n", "\n", "\n", "Variables\n", "\n", "1. battery_power: Energía total que una batería puede almacenar en un tiempo medida en mAh\n", "1. blue: Tiene bluetooth o no\n", "\n", "1. clock_speed: velocidad a la que el microprocesador ejecuta las instrucciones\n", "1. dual_sim: Tiene soporte para dual sim o no\n", "1. fc: Cámara frontal de megapíxeles\n", "1. four_g: Tiene 4G o no\n", "1. int_memory: Memoria interna en gigabytes\n", "1. m_dep: Profundidad del móvil en cm\n", "1. mobile_wt: Peso del móvil\n", "1. n_cores: Número de núcleos del procesador\n", "1. pc: Cámara principal de megapíxeles\n", "1. px_height: Altura de la resolución de píxeles\n", "1. px_width: Ancho de la resolución de píxeles\n", "1. ram: Memoria de acceso aleatorio en Mega Bytes\n", "1. sc_h: Altura de la pantalla del móvil en cm\n", "1. sc_w: Ancho de pantalla del móvil en cm\n", "1. talk_time: tiempo máximo que durará una sola carga de la batería cuando esté\n", "1. three_g: Tiene 3G o no\n", "1. touch_screen: Tiene pantalla táctil o no\n", "1. wifi: Tiene wifi o no\n", "1. Price_range: Esta es la variable objetivo con valor de 0 (costo bajo), 1 (costo medio), 2 (costo alto) y 3 (costo muy alto)." ] }, { "cell_type": "markdown", "id": "804d09a6", "metadata": {}, "source": [ "### Tarea\n", "\n", "Predecir si un celular pertenece alguna de las siguientes clases :\n", "1. 0 (costo bajo)\n", "2. 1 (costo medio)\n", "3. 2 (costo alto)\n", "4. 3 (costo muy alto). " ] }, { "cell_type": "markdown", "id": "c4f50282", "metadata": {}, "source": [ "### 1- Análisis exploratorio de los datos" ] }, { "cell_type": "markdown", "id": "0effc3d9", "metadata": {}, "source": [ "1. Imprima el número de registros del dataset\n", "2. Imprima el número de variables del dataset\n", "3. Imprima el nombre de las columnas del dataset\n", "4. Imprima el **head** del dataset\n", "5. Imprima el **tail** del dataset\n", "6. Imprima **info** basica del dataset\n", "7. Imprima un **describe** del dataset\n", "8. Graficar la distribución de la clase ha predecir (price_range)\n" ] }, { "cell_type": "code", "execution_count": 400, "id": "21d732cd", "metadata": {}, "outputs": [], "source": [ "data = pd.read_csv(\"resources/train.csv\")" ] }, { "cell_type": "code", "execution_count": 402, "id": "823f3323", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Número de regristros 2000\n", "Número de variables 21\n" ] } ], "source": [ "print(\"Número de regristros\",)\n", "print(\"Número de variables\",)" ] }, { "cell_type": "code", "execution_count": 403, "id": "becfbf35", "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", "
battery_powerblueclock_speeddual_simfcfour_gint_memorym_depmobile_wtn_cores...px_heightpx_widthramsc_hsc_wtalk_timethree_gtouch_screenwifiprice_range
084202.201070.61882...20756254997190011
1102110.5101530.71363...9051988263117371102
256310.5121410.91455...12631716260311291102
361512.5000100.81316...121617862769168111002
4182111.20131440.61412...12081212141182151101
\n", "

5 rows × 21 columns

\n", "
" ], "text/plain": [ " battery_power blue clock_speed dual_sim fc four_g int_memory m_dep \\\n", "0 842 0 2.2 0 1 0 7 0.6 \n", "1 1021 1 0.5 1 0 1 53 0.7 \n", "2 563 1 0.5 1 2 1 41 0.9 \n", "3 615 1 2.5 0 0 0 10 0.8 \n", "4 1821 1 1.2 0 13 1 44 0.6 \n", "\n", " mobile_wt n_cores ... px_height px_width ram sc_h sc_w talk_time \\\n", "0 188 2 ... 20 756 2549 9 7 19 \n", "1 136 3 ... 905 1988 2631 17 3 7 \n", "2 145 5 ... 1263 1716 2603 11 2 9 \n", "3 131 6 ... 1216 1786 2769 16 8 11 \n", "4 141 2 ... 1208 1212 1411 8 2 15 \n", "\n", " three_g touch_screen wifi price_range \n", "0 0 0 1 1 \n", "1 1 1 0 2 \n", "2 1 1 0 2 \n", "3 1 0 0 2 \n", "4 1 1 0 1 \n", "\n", "[5 rows x 21 columns]" ] }, "execution_count": 403, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "code", "execution_count": 404, "id": "aa00fa19", "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", "
battery_powerblueclock_speeddual_simfcfour_gint_memorym_depmobile_wtn_cores...px_heightpx_widthramsc_hsc_wtalk_timethree_gtouch_screenwifiprice_range
199579410.510120.81066...12221890668134191100
1996196512.6100390.21874...915196520321110161112
1997191100.9111360.71088...868163230579151103
1998151200.9041460.11455...3366708691810191110
199951012.0151450.91686...483754391919421113
\n", "

5 rows × 21 columns

\n", "
" ], "text/plain": [ " battery_power blue clock_speed dual_sim fc four_g int_memory \\\n", "1995 794 1 0.5 1 0 1 2 \n", "1996 1965 1 2.6 1 0 0 39 \n", "1997 1911 0 0.9 1 1 1 36 \n", "1998 1512 0 0.9 0 4 1 46 \n", "1999 510 1 2.0 1 5 1 45 \n", "\n", " m_dep mobile_wt n_cores ... px_height px_width ram sc_h sc_w \\\n", "1995 0.8 106 6 ... 1222 1890 668 13 4 \n", "1996 0.2 187 4 ... 915 1965 2032 11 10 \n", "1997 0.7 108 8 ... 868 1632 3057 9 1 \n", "1998 0.1 145 5 ... 336 670 869 18 10 \n", "1999 0.9 168 6 ... 483 754 3919 19 4 \n", "\n", " talk_time three_g touch_screen wifi price_range \n", "1995 19 1 1 0 0 \n", "1996 16 1 1 1 2 \n", "1997 5 1 1 0 3 \n", "1998 19 1 1 1 0 \n", "1999 2 1 1 1 3 \n", "\n", "[5 rows x 21 columns]" ] }, "execution_count": 404, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "code", "execution_count": 405, "id": "9edfea18", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RangeIndex: 2000 entries, 0 to 1999\n", "Data columns (total 21 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 battery_power 2000 non-null int64 \n", " 1 blue 2000 non-null int64 \n", " 2 clock_speed 2000 non-null float64\n", " 3 dual_sim 2000 non-null int64 \n", " 4 fc 2000 non-null int64 \n", " 5 four_g 2000 non-null int64 \n", " 6 int_memory 2000 non-null int64 \n", " 7 m_dep 2000 non-null float64\n", " 8 mobile_wt 2000 non-null int64 \n", " 9 n_cores 2000 non-null int64 \n", " 10 pc 2000 non-null int64 \n", " 11 px_height 2000 non-null int64 \n", " 12 px_width 2000 non-null int64 \n", " 13 ram 2000 non-null int64 \n", " 14 sc_h 2000 non-null int64 \n", " 15 sc_w 2000 non-null int64 \n", " 16 talk_time 2000 non-null int64 \n", " 17 three_g 2000 non-null int64 \n", " 18 touch_screen 2000 non-null int64 \n", " 19 wifi 2000 non-null int64 \n", " 20 price_range 2000 non-null int64 \n", "dtypes: float64(2), int64(19)\n", "memory usage: 328.2 KB\n" ] } ], "source": [] }, { "cell_type": "code", "execution_count": 406, "id": "0fb51ec5", "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", "
battery_powerblueclock_speeddual_simfcfour_gint_memorym_depmobile_wtn_cores...px_heightpx_widthramsc_hsc_wtalk_timethree_gtouch_screenwifiprice_range
count2000.0000002000.00002000.0000002000.0000002000.0000002000.0000002000.0000002000.0000002000.0000002000.000000...2000.0000002000.0000002000.0000002000.0000002000.0000002000.0000002000.0000002000.0000002000.0000002000.000000
mean1238.5185000.49501.5222500.5095004.3095000.52150032.0465000.501750140.2490004.520500...645.1080001251.5155002124.21300012.3065005.76700011.0110000.7615000.5030000.5070001.500000
std439.4182060.50010.8160040.5000354.3414440.49966218.1457150.28841635.3996552.287837...443.780811432.1994471084.7320444.2132454.3563985.4639550.4262730.5001160.5000761.118314
min501.0000000.00000.5000000.0000000.0000000.0000002.0000000.10000080.0000001.000000...0.000000500.000000256.0000005.0000000.0000002.0000000.0000000.0000000.0000000.000000
25%851.7500000.00000.7000000.0000001.0000000.00000016.0000000.200000109.0000003.000000...282.750000874.7500001207.5000009.0000002.0000006.0000001.0000000.0000000.0000000.750000
50%1226.0000000.00001.5000001.0000003.0000001.00000032.0000000.500000141.0000004.000000...564.0000001247.0000002146.50000012.0000005.00000011.0000001.0000001.0000001.0000001.500000
75%1615.2500001.00002.2000001.0000007.0000001.00000048.0000000.800000170.0000007.000000...947.2500001633.0000003064.50000016.0000009.00000016.0000001.0000001.0000001.0000002.250000
max1998.0000001.00003.0000001.00000019.0000001.00000064.0000001.000000200.0000008.000000...1960.0000001998.0000003998.00000019.00000018.00000020.0000001.0000001.0000001.0000003.000000
\n", "

8 rows × 21 columns

\n", "
" ], "text/plain": [ " battery_power blue clock_speed dual_sim fc \\\n", "count 2000.000000 2000.0000 2000.000000 2000.000000 2000.000000 \n", "mean 1238.518500 0.4950 1.522250 0.509500 4.309500 \n", "std 439.418206 0.5001 0.816004 0.500035 4.341444 \n", "min 501.000000 0.0000 0.500000 0.000000 0.000000 \n", "25% 851.750000 0.0000 0.700000 0.000000 1.000000 \n", "50% 1226.000000 0.0000 1.500000 1.000000 3.000000 \n", "75% 1615.250000 1.0000 2.200000 1.000000 7.000000 \n", "max 1998.000000 1.0000 3.000000 1.000000 19.000000 \n", "\n", " four_g int_memory m_dep mobile_wt n_cores ... \\\n", "count 2000.000000 2000.000000 2000.000000 2000.000000 2000.000000 ... \n", "mean 0.521500 32.046500 0.501750 140.249000 4.520500 ... \n", "std 0.499662 18.145715 0.288416 35.399655 2.287837 ... \n", "min 0.000000 2.000000 0.100000 80.000000 1.000000 ... \n", "25% 0.000000 16.000000 0.200000 109.000000 3.000000 ... \n", "50% 1.000000 32.000000 0.500000 141.000000 4.000000 ... \n", "75% 1.000000 48.000000 0.800000 170.000000 7.000000 ... \n", "max 1.000000 64.000000 1.000000 200.000000 8.000000 ... \n", "\n", " px_height px_width ram sc_h sc_w \\\n", "count 2000.000000 2000.000000 2000.000000 2000.000000 2000.000000 \n", "mean 645.108000 1251.515500 2124.213000 12.306500 5.767000 \n", "std 443.780811 432.199447 1084.732044 4.213245 4.356398 \n", "min 0.000000 500.000000 256.000000 5.000000 0.000000 \n", "25% 282.750000 874.750000 1207.500000 9.000000 2.000000 \n", "50% 564.000000 1247.000000 2146.500000 12.000000 5.000000 \n", "75% 947.250000 1633.000000 3064.500000 16.000000 9.000000 \n", "max 1960.000000 1998.000000 3998.000000 19.000000 18.000000 \n", "\n", " talk_time three_g touch_screen wifi price_range \n", "count 2000.000000 2000.000000 2000.000000 2000.000000 2000.000000 \n", "mean 11.011000 0.761500 0.503000 0.507000 1.500000 \n", "std 5.463955 0.426273 0.500116 0.500076 1.118314 \n", "min 2.000000 0.000000 0.000000 0.000000 0.000000 \n", "25% 6.000000 1.000000 0.000000 0.000000 0.750000 \n", "50% 11.000000 1.000000 1.000000 1.000000 1.500000 \n", "75% 16.000000 1.000000 1.000000 1.000000 2.250000 \n", "max 20.000000 1.000000 1.000000 1.000000 3.000000 \n", "\n", "[8 rows x 21 columns]" ] }, "execution_count": 406, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "code", "execution_count": 407, "id": "7fba415a", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([1, 2, 3, 0], dtype=int64)" ] }, "execution_count": 407, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "code", "execution_count": 408, "id": "4b7e9309", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\seaborn\\_decorators.py:36: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n", " warnings.warn(\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 408, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [] }, { "cell_type": "markdown", "id": "fb679511", "metadata": {}, "source": [ "9. Graficar la distribución de los n_cores\n" ] }, { "cell_type": "code", "execution_count": 412, "id": "8f32d035", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 412, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [] }, { "cell_type": "markdown", "id": "cdf5a6a3", "metadata": {}, "source": [ "### 2. Tratamiento de missing, reparación dataset y codificación de variables" ] }, { "cell_type": "code", "execution_count": 681, "id": "50eb9f51", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "f5afdd5a", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 684, "id": "c5cb1a97", "metadata": {}, "outputs": [], "source": [ "scaler = StandardScaler()" ] }, { "cell_type": "code", "execution_count": null, "id": "8635441d", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "id": "7fbb4302", "metadata": {}, "source": [ "### 3. Determinar el conjunto de entrenamiento y el de validación.\n", "\n", "1. Hacer división de los datos 80% train , 20% test Crear un vector X el cual contiene las características \n", "2. Imprimir el shape o dimensiones del vector de entrenamiento (x_train)\n", "2. Imprimir el shape o dimensiones del vector de prueba (x_test)\n", "Ayuda: usar la función train_test_split de sklearn https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html\n", "\n" ] }, { "cell_type": "code", "execution_count": 752, "id": "15bafdd7", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 753, "id": "1a6bb765", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Dimensiones vector de entrenamiento (1600, 20)\n" ] } ], "source": [ "print(\"Dimensiones vector de entrenamiento\", )" ] }, { "cell_type": "code", "execution_count": 754, "id": "a8209442", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Dimensiones vector de prueba (400, 20)\n" ] } ], "source": [ "print(\"Dimensiones vector de prueba\", )" ] }, { "cell_type": "code", "execution_count": 755, "id": "4f20d139", "metadata": {}, "outputs": [], "source": [ "x_train = scaler.fit_transform(x_train) #Normalizamos los datos" ] }, { "cell_type": "markdown", "id": "4f345f89", "metadata": {}, "source": [ "### 4. Entrenamiento del modelo\n", "\n", "\n", "1. Crear un MLPClassifier model usando la librería sklearn https://scikit-learn.org/stable/modules/generated/sklearn.neural_network.MLPClassifier.html\n", "2. Entrenar el modelo\n", "3. Usar 4 capas ocultas de (5000,1000,1000,10) neuronas\n", "5. Usar early_stopping = True\n", "4. Usar un alpha=1e-5 \n", "6. Usar solver='lbfgs'\n", "\n", "Ayudas:\n", "\n", "- Usar la función fit\n", "- Solo usar el conjunto de entrenamiento (X_train, y_train)" ] }, { "cell_type": "code", "execution_count": 798, "id": "d333355e", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 799, "id": "782040b7", "metadata": { "scrolled": false }, "outputs": [ { "data": { "text/plain": [ "MLPClassifier(alpha=1e-05, early_stopping=True,\n", " hidden_layer_sizes=(5000, 1000, 1000, 10), random_state=1,\n", " solver='lbfgs', verbose=True, warm_start=True)" ] }, "execution_count": 799, "metadata": {}, "output_type": "execute_result" } ], "source": [] }, { "cell_type": "code", "execution_count": 800, "id": "c33c43b4", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Número de iteraciones necesarias para entrenar el modelo 51\n" ] } ], "source": [ "print(\"Número de iteraciones necesarias para entrenar el modelo\",)" ] }, { "cell_type": "markdown", "id": "68998628", "metadata": {}, "source": [ "### 5. Calcular las métricas de evaluación\n", "\n", "**Nota:** Ejecutar la siguiente función, la cual calcula crea la matriz de confusión y algunas métricas. " ] }, { "cell_type": "code", "execution_count": 801, "id": "29a5fb36", "metadata": {}, "outputs": [], "source": [ " def metrics(y_true,y_pred):\n", " \"\"\"\n", " This method calculate some metrics shuch as acurracy,f1-score,precision and create confusion matrix figure.\n", "\n", " Args:\n", " y_true (numpy_array): true classes\n", " y_pred (numpy_array): predict classes\n", "\n", " Returns:\n", " \n", " cm_fig (ConfusionMatrixDisplay: Confusion matrix figure\n", " accuracy (float): acurracy\n", " report (dict): some metrics\n", "\n", " \"\"\"\n", " cm = confusion_matrix(y_true,y_pred, normalize='true')\n", " report = classification_report(y_true,y_pred,output_dict=True)\n", " cm_fig = ConfusionMatrixDisplay(confusion_matrix=cm)\n", " return cm_fig,report[\"accuracy\"],report" ] }, { "cell_type": "markdown", "id": "6be37a3d", "metadata": {}, "source": [ "1. Usar la función predict() para crear el vector de predicciones\n", "\n", "\n", "Ayuda: Utilice el conjunto de test (X_test)" ] }, { "cell_type": "code", "execution_count": 802, "id": "a65904f2", "metadata": {}, "outputs": [], "source": [ "x_test_norma = scaler.transform(x_test)" ] }, { "cell_type": "code", "execution_count": 803, "id": "8c99b503", "metadata": {}, "outputs": [], "source": [ "y_predict = clf.predict(x_test_norma)" ] }, { "cell_type": "code", "execution_count": 804, "id": "b3babd3a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "ACCURACY 0.95\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "\"\"\"\n", "Utiliza la función metrics, debes reemplazar las variables\n", "y_test por las clases del conjunto de test y y_predict por las predicciones obtenidas de tu modelo.\n", "\n", "\"\"\"\n", "cm_fig,test_score, report = metrics(y_test,y_predict)\n", "cm_fig.plot(cmap=plt.cm.Blues)\n", "print(\"ACCURACY\",test_score)" ] }, { "cell_type": "markdown", "id": "9f29571a", "metadata": {}, "source": [ "### 6. Conclusiones\n", "\n", "Describa brevemente los resultados obtenidos, incluyendo el accuracy y mencionando el comportamiento del modelo clasificando muestras para ambas clases." ] }, { "cell_type": "markdown", "id": "b04952a7", "metadata": {}, "source": [ "Como podemos ver el modelo es realmente bueno ya que con muestras que jamás conoció durante el entrenamiento obtuvo un accuracy de 0.95 además tiene un buen resultado clasificando todas las clases:\n", "\n", "- 0 (costo bajo)\n", "- 1 (costo medio)\n", "- 2 (costo alto)\n", "- 3 (costo muy alto)." ] }, { "cell_type": "markdown", "id": "3f4d3c16", "metadata": {}, "source": [ "\n", "\n", "\n", "Profesor: Jose Alberto Arango Sánchez
[](https://www.linkedin.com/in/jose-alberto-arango-sanchez-79a337142/)\n", "\n", " \n", "\n", "@jose.arangos
[](https://github.com/josearangos)" ] }, { "cell_type": "markdown", "id": "b41d532b", "metadata": {}, "source": [ "" ] } ], "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": "3.9.12" } }, "nbformat": 4, "nbformat_minor": 5 }