{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "61ef78a3", "metadata": {}, "outputs": [], "source": [ "from pycaret.classification import *\n", "from pycaret.datasets import get_data\n", "import pandas as pd" ] }, { "cell_type": "markdown", "id": "a5e8eefc", "metadata": {}, "source": [ "## Getting the dataset" ] }, { "cell_type": "code", "execution_count": 2, "id": "7e484728", "metadata": { "scrolled": false }, "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", "
agejobmaritaleducationdefaultbalancehousingloancontactdaymonthdurationcampaignpdayspreviouspoutcomedeposit
058managementmarriedtertiaryno2143yesnounknown5may2611-10unknownno
144techniciansinglesecondaryno29yesnounknown5may1511-10unknownno
233entrepreneurmarriedsecondaryno2yesyesunknown5may761-10unknownno
347blue-collarmarriedunknownno1506yesnounknown5may921-10unknownno
433unknownsingleunknownno1nonounknown5may1981-10unknownno
\n", "
" ], "text/plain": [ " age job marital education default balance housing loan \\\n", "0 58 management married tertiary no 2143 yes no \n", "1 44 technician single secondary no 29 yes no \n", "2 33 entrepreneur married secondary no 2 yes yes \n", "3 47 blue-collar married unknown no 1506 yes no \n", "4 33 unknown single unknown no 1 no no \n", "\n", " contact day month duration campaign pdays previous poutcome deposit \n", "0 unknown 5 may 261 1 -1 0 unknown no \n", "1 unknown 5 may 151 1 -1 0 unknown no \n", "2 unknown 5 may 76 1 -1 0 unknown no \n", "3 unknown 5 may 92 1 -1 0 unknown no \n", "4 unknown 5 may 198 1 -1 0 unknown no " ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "#get dataset\n", "dataset = get_data('bank')" ] }, { "cell_type": "code", "execution_count": 3, "id": "c9db90e0", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(45211, 17)" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "#check the shape of data\n", "dataset.shape" ] }, { "cell_type": "code", "execution_count": 5, "id": "0fb2e781", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "age 0\n", "job 0\n", "marital 0\n", "education 0\n", "default 0\n", "balance 0\n", "housing 0\n", "loan 0\n", "contact 0\n", "day 0\n", "month 0\n", "duration 0\n", "campaign 0\n", "pdays 0\n", "previous 0\n", "poutcome 0\n", "deposit 0\n", "dtype: int64" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "dataset.isna().sum()" ] }, { "cell_type": "code", "execution_count": 6, "id": "a18b10c3", "metadata": {}, "outputs": [], "source": [ "## sample returns a random sample from an axis of the object. That would be 38,429 samples, not 45211\n", "data = dataset.sample(frac=0.85, random_state=456)" ] }, { "cell_type": "code", "execution_count": 7, "id": "cb03c457", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
agejobmaritaleducationdefaultbalancehousingloancontactdaymonthdurationcampaignpdayspreviouspoutcomedeposit
886329blue-collarmarriedprimaryno25yesnounknown4jun1882-10unknownno
2268840techniciandivorcedsecondaryno237nonocellular25aug875-10unknownno
96157retiredmarriedtertiaryno906yesnounknown7may1171-10unknownno
1022945servicesmarriedprimaryno116yesnounknown11jun2873-10unknownno
2118948blue-collarmarriedprimaryno-83nonocellular14aug1363-10unknownno
......................................................
2572858blue-collardivorcedprimaryno8218yesnocellular19nov141211110failureno
3343028studentsinglesecondaryno0nonocellular20apr1851-10unknownyes
748144blue-collarsingleprimaryno1593yesnounknown29may8283-10unknownyes
459340servicesmarriedprimaryno3559yesnounknown20may1388-10unknownno
4268037managementmarriedtertiaryno0nonocellular15jan42621961otheryes
\n", "

38429 rows × 17 columns

\n", "
" ], "text/plain": [ " age job marital education default balance housing loan \\\n", "8863 29 blue-collar married primary no 25 yes no \n", "22688 40 technician divorced secondary no 237 no no \n", "961 57 retired married tertiary no 906 yes no \n", "10229 45 services married primary no 116 yes no \n", "21189 48 blue-collar married primary no -83 no no \n", "... ... ... ... ... ... ... ... ... \n", "25728 58 blue-collar divorced primary no 8218 yes no \n", "33430 28 student single secondary no 0 no no \n", "7481 44 blue-collar single primary no 1593 yes no \n", "4593 40 services married primary no 3559 yes no \n", "42680 37 management married tertiary no 0 no no \n", "\n", " contact day month duration campaign pdays previous poutcome \\\n", "8863 unknown 4 jun 188 2 -1 0 unknown \n", "22688 cellular 25 aug 87 5 -1 0 unknown \n", "961 unknown 7 may 117 1 -1 0 unknown \n", "10229 unknown 11 jun 287 3 -1 0 unknown \n", "21189 cellular 14 aug 136 3 -1 0 unknown \n", "... ... ... ... ... ... ... ... ... \n", "25728 cellular 19 nov 141 2 111 10 failure \n", "33430 cellular 20 apr 185 1 -1 0 unknown \n", "7481 unknown 29 may 828 3 -1 0 unknown \n", "4593 unknown 20 may 138 8 -1 0 unknown \n", "42680 cellular 15 jan 426 2 196 1 other \n", "\n", " deposit \n", "8863 no \n", "22688 no \n", "961 no \n", "10229 no \n", "21189 no \n", "... ... \n", "25728 no \n", "33430 yes \n", "7481 yes \n", "4593 no \n", "42680 yes \n", "\n", "[38429 rows x 17 columns]" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data" ] }, { "cell_type": "code", "execution_count": 8, "id": "35afe966", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
agejobmaritaleducationdefaultbalancehousingloancontactdaymonthdurationcampaignpdayspreviouspoutcomedeposit
233entrepreneurmarriedsecondaryno2yesyesunknown5may761-10unknownno
433unknownsingleunknownno1nonounknown5may1981-10unknownno
1645admin.singleunknownno13yesnounknown5may981-10unknownno
4150managementmarriedsecondaryno49yesnounknown5may1802-10unknownno
5232managementmarriedtertiaryno0yesnounknown5may1791-10unknownno
......................................................
4518463retiredmarriedsecondaryno1495nonocellular16nov1381225successno
4518925servicessinglesecondaryno199nonocellular16nov1731925failureno
4520153managementmarriedtertiaryno583nonocellular17nov22611844successyes
4520771retireddivorcedprimaryno1729nonocellular17nov4562-10unknownyes
4521037entrepreneurmarriedsecondaryno2971nonocellular17nov361218811otherno
\n", "

6782 rows × 17 columns

\n", "
" ], "text/plain": [ " age job marital education default balance housing loan \\\n", "2 33 entrepreneur married secondary no 2 yes yes \n", "4 33 unknown single unknown no 1 no no \n", "16 45 admin. single unknown no 13 yes no \n", "41 50 management married secondary no 49 yes no \n", "52 32 management married tertiary no 0 yes no \n", "... ... ... ... ... ... ... ... ... \n", "45184 63 retired married secondary no 1495 no no \n", "45189 25 services single secondary no 199 no no \n", "45201 53 management married tertiary no 583 no no \n", "45207 71 retired divorced primary no 1729 no no \n", "45210 37 entrepreneur married secondary no 2971 no no \n", "\n", " contact day month duration campaign pdays previous poutcome \\\n", "2 unknown 5 may 76 1 -1 0 unknown \n", "4 unknown 5 may 198 1 -1 0 unknown \n", "16 unknown 5 may 98 1 -1 0 unknown \n", "41 unknown 5 may 180 2 -1 0 unknown \n", "52 unknown 5 may 179 1 -1 0 unknown \n", "... ... ... ... ... ... ... ... ... \n", "45184 cellular 16 nov 138 1 22 5 success \n", "45189 cellular 16 nov 173 1 92 5 failure \n", "45201 cellular 17 nov 226 1 184 4 success \n", "45207 cellular 17 nov 456 2 -1 0 unknown \n", "45210 cellular 17 nov 361 2 188 11 other \n", "\n", " deposit \n", "2 no \n", "4 no \n", "16 no \n", "41 no \n", "52 no \n", "... ... \n", "45184 no \n", "45189 no \n", "45201 yes \n", "45207 yes \n", "45210 no \n", "\n", "[6782 rows x 17 columns]" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# remove from the original dataset this random data\n", "data_unseen = dataset.drop(data.index)\n", "data_unseen" ] }, { "cell_type": "code", "execution_count": 9, "id": "31f3169f", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Data for Modeling: (38429, 17)\n", "Unseen Data For Predictions: (6782, 17)\n" ] } ], "source": [ "# Reseting the index of both datasets\n", "data.reset_index(inplace=True, drop=True)\n", "data_unseen.reset_index(inplace=True, drop=True)\n", "print('Data for Modeling: ' + str(data.shape))\n", "print('Unseen Data For Predictions: ' + str(data_unseen.shape))" ] }, { "cell_type": "code", "execution_count": 10, "id": "8fa0278a", "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", "
 DescriptionValue
0Session id321
1Targetdeposit
2Target typeBinary
3Target mappingno: 0, yes: 1
4Original data shape(38429, 17)
5Transformed data shape(38429, 49)
6Transformed train set shape(26900, 49)
7Transformed test set shape(11529, 49)
8Ordinal features3
9Numeric features7
10Categorical features9
11PreprocessTrue
12Imputation typesimple
13Numeric imputationmean
14Categorical imputationmode
15Maximum one-hot encoding25
16Encoding methodNone
17Fold GeneratorStratifiedKFold
18Fold Number10
19CPU Jobs-1
20Use GPUFalse
21Log ExperimentFalse
22Experiment Nameclf-default-name
23USIdfd7
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "model_setup = setup(data=data, target='deposit', session_id=321)" ] }, { "cell_type": "markdown", "id": "a2e17f35", "metadata": {}, "source": [ "## Compare Model" ] }, { "cell_type": "code", "execution_count": 11, "id": "c5ff181d", "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "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", "
 ModelAccuracyAUCRecallPrec.F1KappaMCCTT (Sec)
catboostCatBoost Classifier0.90910.93640.48390.64950.55410.50470.51181.4950
lightgbmLight Gradient Boosting Machine0.90810.93470.48140.64320.54990.50000.50691.4960
xgboostExtreme Gradient Boosting0.90720.93090.48840.63440.55130.50050.50621.2350
gbcGradient Boosting Classifier0.90560.92620.40540.65560.50030.45140.46771.4290
rfRandom Forest Classifier0.90390.92670.36530.66220.47030.42230.44531.5270
lrLogistic Regression0.90150.90380.34750.64590.45100.40230.42612.5670
ldaLinear Discriminant Analysis0.90020.90780.44320.59870.50890.45480.46141.2640
ridgeRidge Classifier0.90000.00000.28380.67160.39820.35340.39311.4140
adaAda Boost Classifier0.89970.90930.37990.61430.46860.41680.43191.3580
etExtra Trees Classifier0.89840.90500.32480.62920.42790.37830.40361.6490
dummyDummy Classifier0.88320.50000.00000.00000.00000.00000.00001.2130
knnK Neighbors Classifier0.88190.75840.26440.48990.34320.28470.30122.0930
dtDecision Tree Classifier0.87360.70370.48200.46110.47090.39920.39961.4360
nbNaive Bayes0.85970.82500.52150.41960.46490.38530.38841.3590
qdaQuadratic Discriminant Analysis0.85860.82310.47730.42400.44300.36400.36791.4850
svmSVM - Linear Kernel0.83020.00000.20810.26740.20540.12240.13471.4500
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Processing: 0%| | 0/69 [00:00\n" ] } ], "source": [ "print(best_model)" ] }, { "cell_type": "markdown", "id": "bddc36a6", "metadata": {}, "source": [ "## Create the Model" ] }, { "cell_type": "code", "execution_count": 13, "id": "b3dcb641", "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", "
NameReferenceTurbo
ID
lrLogistic Regressionsklearn.linear_model._logistic.LogisticRegressionTrue
knnK Neighbors Classifiersklearn.neighbors._classification.KNeighborsCl...True
nbNaive Bayessklearn.naive_bayes.GaussianNBTrue
dtDecision Tree Classifiersklearn.tree._classes.DecisionTreeClassifierTrue
svmSVM - Linear Kernelsklearn.linear_model._stochastic_gradient.SGDC...True
rbfsvmSVM - Radial Kernelsklearn.svm._classes.SVCFalse
gpcGaussian Process Classifiersklearn.gaussian_process._gpc.GaussianProcessC...False
mlpMLP Classifiersklearn.neural_network._multilayer_perceptron....False
ridgeRidge Classifiersklearn.linear_model._ridge.RidgeClassifierTrue
rfRandom Forest Classifiersklearn.ensemble._forest.RandomForestClassifierTrue
qdaQuadratic Discriminant Analysissklearn.discriminant_analysis.QuadraticDiscrim...True
adaAda Boost Classifiersklearn.ensemble._weight_boosting.AdaBoostClas...True
gbcGradient Boosting Classifiersklearn.ensemble._gb.GradientBoostingClassifierTrue
ldaLinear Discriminant Analysissklearn.discriminant_analysis.LinearDiscrimina...True
etExtra Trees Classifiersklearn.ensemble._forest.ExtraTreesClassifierTrue
xgboostExtreme Gradient Boostingxgboost.sklearn.XGBClassifierTrue
lightgbmLight Gradient Boosting Machinelightgbm.sklearn.LGBMClassifierTrue
catboostCatBoost Classifiercatboost.core.CatBoostClassifierTrue
dummyDummy Classifiersklearn.dummy.DummyClassifierTrue
\n", "
" ], "text/plain": [ " Name \\\n", "ID \n", "lr Logistic Regression \n", "knn K Neighbors Classifier \n", "nb Naive Bayes \n", "dt Decision Tree Classifier \n", "svm SVM - Linear Kernel \n", "rbfsvm SVM - Radial Kernel \n", "gpc Gaussian Process Classifier \n", "mlp MLP Classifier \n", "ridge Ridge Classifier \n", "rf Random Forest Classifier \n", "qda Quadratic Discriminant Analysis \n", "ada Ada Boost Classifier \n", "gbc Gradient Boosting Classifier \n", "lda Linear Discriminant Analysis \n", "et Extra Trees Classifier \n", "xgboost Extreme Gradient Boosting \n", "lightgbm Light Gradient Boosting Machine \n", "catboost CatBoost Classifier \n", "dummy Dummy Classifier \n", "\n", " Reference Turbo \n", "ID \n", "lr sklearn.linear_model._logistic.LogisticRegression True \n", "knn sklearn.neighbors._classification.KNeighborsCl... True \n", "nb sklearn.naive_bayes.GaussianNB True \n", "dt sklearn.tree._classes.DecisionTreeClassifier True \n", "svm sklearn.linear_model._stochastic_gradient.SGDC... True \n", "rbfsvm sklearn.svm._classes.SVC False \n", "gpc sklearn.gaussian_process._gpc.GaussianProcessC... False \n", "mlp sklearn.neural_network._multilayer_perceptron.... False \n", "ridge sklearn.linear_model._ridge.RidgeClassifier True \n", "rf sklearn.ensemble._forest.RandomForestClassifier True \n", "qda sklearn.discriminant_analysis.QuadraticDiscrim... True \n", "ada sklearn.ensemble._weight_boosting.AdaBoostClas... True \n", "gbc sklearn.ensemble._gb.GradientBoostingClassifier True \n", "lda sklearn.discriminant_analysis.LinearDiscrimina... True \n", "et sklearn.ensemble._forest.ExtraTreesClassifier True \n", "xgboost xgboost.sklearn.XGBClassifier True \n", "lightgbm lightgbm.sklearn.LGBMClassifier True \n", "catboost catboost.core.CatBoostClassifier True \n", "dummy sklearn.dummy.DummyClassifier True " ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "models()" ] }, { "cell_type": "code", "execution_count": 14, "id": "90173a54", "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "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", "
 AccuracyAUCRecallPrec.F1KappaMCC
Fold       
00.91640.93360.48410.70700.57470.53010.5418
10.90780.92760.38540.68750.49390.44760.4703
20.90040.92530.36620.62500.46180.41100.4289
30.90630.92070.42360.65200.51350.46430.4775
40.90780.93690.41080.67190.50990.46220.4793
50.90590.92680.40760.65640.50290.45410.4699
60.90110.93090.42360.61010.50000.44710.4563
70.90480.92210.37140.66860.47760.42990.4524
80.90110.91400.36510.63540.46370.41360.4329
90.90450.92400.41590.64220.50480.45460.4678
Mean0.90560.92620.40540.65560.50030.45140.4677
Std0.00440.00630.03420.02780.03030.03170.0297
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Processing: 0%| | 0/4 [00:00" ] }, "metadata": {}, "output_type": "display_data" }, { "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", "
 AccuracyAUCRecallPrec.F1KappaMCC
Fold       
00.91040.93730.48410.65800.55780.50920.5167
10.90820.92950.42990.66500.52220.47400.4879
20.90450.92780.48410.61540.54190.48940.4938
30.90970.92530.51590.64030.57140.52160.5254
40.91000.94180.47130.66070.55020.50180.5106
50.90410.92410.46820.61760.53260.48030.4860
60.90590.93260.49040.62350.54900.49740.5018
70.90370.92880.46030.61970.52820.47590.4824
80.90110.92070.45400.60340.51810.46420.4701
90.90710.92830.48570.63490.55040.49960.5051
Mean0.90650.92960.47440.63380.54220.49130.4980
Std0.00300.00590.02220.02040.01590.01690.0162
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Processing: 0%| | 0/7 [00:00" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "## AUC Plot\n", "\n", "plot_model(tuned_gbc, plot = 'auc')\n" ] }, { "cell_type": "code", "execution_count": 20, "id": "92a67823", "metadata": {}, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "## Consufion matrix\n", "\n", "plot_model(tuned_gbc, plot = 'confusion_matrix')" ] }, { "cell_type": "markdown", "id": "bda74342", "metadata": {}, "source": [ "## Evaluation the model\n" ] }, { "cell_type": "code", "execution_count": 21, "id": "00249ace", "metadata": { "scrolled": true }, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "53ad94081d7142a9948bf6bb6cfb7d1d", "version_major": 2, "version_minor": 0 }, "text/plain": [ "interactive(children=(ToggleButtons(description='Plot Type:', icons=('',), options=(('Pipeline Plot', 'pipelin…" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "## model performance is to use the evaluate_model()\n", "evaluate_model(tuned_gbc)" ] }, { "cell_type": "markdown", "id": "ebaf37a8", "metadata": {}, "source": [ "## Finalizing the Model" ] }, { "cell_type": "code", "execution_count": 22, "id": "ae1c129c", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
Pipeline(memory=FastMemory(location=C:\\Users\\owner\\AppData\\Local\\Temp\\joblib),\n",
       "         steps=[('label_encoding',\n",
       "                 TransformerWrapperWithInverse(exclude=None, include=None,\n",
       "                                               transformer=LabelEncoder())),\n",
       "                ('numerical_imputer',\n",
       "                 TransformerWrapper(exclude=None,\n",
       "                                    include=['age', 'balance', 'day',\n",
       "                                             'duration', 'campaign', 'pdays',\n",
       "                                             'previous'],\n",
       "                                    transformer=SimpleImputer(add_indica...\n",
       "                                            criterion='friedman_mse', init=None,\n",
       "                                            learning_rate=0.3, loss='log_loss',\n",
       "                                            max_depth=5, max_features='sqrt',\n",
       "                                            max_leaf_nodes=None,\n",
       "                                            min_impurity_decrease=0.01,\n",
       "                                            min_samples_leaf=3,\n",
       "                                            min_samples_split=4,\n",
       "                                            min_weight_fraction_leaf=0.0,\n",
       "                                            n_estimators=80,\n",
       "                                            n_iter_no_change=None,\n",
       "                                            random_state=321, subsample=0.95,\n",
       "                                            tol=0.0001, validation_fraction=0.1,\n",
       "                                            verbose=0, warm_start=False))],\n",
       "         verbose=False)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "Pipeline(memory=FastMemory(location=C:\\Users\\owner\\AppData\\Local\\Temp\\joblib),\n", " steps=[('label_encoding',\n", " TransformerWrapperWithInverse(exclude=None, include=None,\n", " transformer=LabelEncoder())),\n", " ('numerical_imputer',\n", " TransformerWrapper(exclude=None,\n", " include=['age', 'balance', 'day',\n", " 'duration', 'campaign', 'pdays',\n", " 'previous'],\n", " transformer=SimpleImputer(add_indica...\n", " criterion='friedman_mse', init=None,\n", " learning_rate=0.3, loss='log_loss',\n", " max_depth=5, max_features='sqrt',\n", " max_leaf_nodes=None,\n", " min_impurity_decrease=0.01,\n", " min_samples_leaf=3,\n", " min_samples_split=4,\n", " min_weight_fraction_leaf=0.0,\n", " n_estimators=80,\n", " n_iter_no_change=None,\n", " random_state=321, subsample=0.95,\n", " tol=0.0001, validation_fraction=0.1,\n", " verbose=0, warm_start=False))],\n", " verbose=False)" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "final_gbc = finalize_model(tuned_gbc)\n", "final_gbc" ] }, { "cell_type": "code", "execution_count": 23, "id": "7a4bcc6e", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Pipeline(memory=FastMemory(location=C:\\Users\\owner\\AppData\\Local\\Temp\\joblib),\n", " steps=[('label_encoding',\n", " TransformerWrapperWithInverse(exclude=None, include=None,\n", " transformer=LabelEncoder())),\n", " ('numerical_imputer',\n", " TransformerWrapper(exclude=None,\n", " include=['age', 'balance', 'day',\n", " 'duration', 'campaign', 'pdays',\n", " 'previous'],\n", " transformer=SimpleImputer(add_indica...\n", " criterion='friedman_mse', init=None,\n", " learning_rate=0.3, loss='log_loss',\n", " max_depth=5, max_features='sqrt',\n", " max_leaf_nodes=None,\n", " min_impurity_decrease=0.01,\n", " min_samples_leaf=3,\n", " min_samples_split=4,\n", " min_weight_fraction_leaf=0.0,\n", " n_estimators=80,\n", " n_iter_no_change=None,\n", " random_state=321, subsample=0.95,\n", " tol=0.0001, validation_fraction=0.1,\n", " verbose=0, warm_start=False))],\n", " verbose=False)\n" ] } ], "source": [ "#Final gbc parameters for deployment\n", "print(final_gbc)" ] }, { "cell_type": "markdown", "id": "36670c58", "metadata": {}, "source": [ "## Predicting with the model" ] }, { "cell_type": "code", "execution_count": 24, "id": "1295a1e6", "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", "
 ModelAccuracyAUCRecallPrec.F1KappaMCC
0Gradient Boosting Classifier0.93000.95630000.62170.6289
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
agejobmaritaleducationdefaultbalancehousingloancontactdaymonthdurationcampaignpdayspreviouspoutcomedepositprediction_labelprediction_score
652842managementmarriedtertiaryno144nonocellular4mar1481874failureyesno0.5555
3403251unemployedmarriedsecondaryno636nonocellular30jan3211-10unknownnono0.9481
3075436blue-collarsinglesecondaryno2235yesnocellular20nov2872-10unknownnono0.9765
3445648servicesmarriedsecondaryno116yesnotelephone20apr704-10unknownnono0.9978
1345930unknownsingletertiaryno6836nonocellular27feb303-10unknownnono0.7563
............................................................
572856servicesmarriedsecondaryno83nonocellular27aug2611-10unknownnono0.9970
2552042blue-collardivorcedunknownno0nonocellular7jul642-10unknownnono0.9964
2623231servicesmarriedsecondaryno428yesnounknown21may2721-10unknownnono0.9913
1943430blue-collarmarriedsecondaryno664noyestelephone14may571-10unknownnono0.9900
3019255managementmarriedtertiaryno236nonocellular4aug2001-10unknownnono0.9249
\n", "

11529 rows × 19 columns

\n", "
" ], "text/plain": [ " age job marital education default balance housing loan \\\n", "6528 42 management married tertiary no 144 no no \n", "34032 51 unemployed married secondary no 636 no no \n", "30754 36 blue-collar single secondary no 2235 yes no \n", "34456 48 services married secondary no 116 yes no \n", "13459 30 unknown single tertiary no 6836 no no \n", "... ... ... ... ... ... ... ... ... \n", "5728 56 services married secondary no 83 no no \n", "25520 42 blue-collar divorced unknown no 0 no no \n", "26232 31 services married secondary no 428 yes no \n", "19434 30 blue-collar married secondary no 664 no yes \n", "30192 55 management married tertiary no 236 no no \n", "\n", " contact day month duration campaign pdays previous poutcome \\\n", "6528 cellular 4 mar 148 1 87 4 failure \n", "34032 cellular 30 jan 321 1 -1 0 unknown \n", "30754 cellular 20 nov 287 2 -1 0 unknown \n", "34456 telephone 20 apr 70 4 -1 0 unknown \n", "13459 cellular 27 feb 30 3 -1 0 unknown \n", "... ... ... ... ... ... ... ... ... \n", "5728 cellular 27 aug 26 11 -1 0 unknown \n", "25520 cellular 7 jul 64 2 -1 0 unknown \n", "26232 unknown 21 may 272 1 -1 0 unknown \n", "19434 telephone 14 may 57 1 -1 0 unknown \n", "30192 cellular 4 aug 200 1 -1 0 unknown \n", "\n", " deposit prediction_label prediction_score \n", "6528 yes no 0.5555 \n", "34032 no no 0.9481 \n", "30754 no no 0.9765 \n", "34456 no no 0.9978 \n", "13459 no no 0.7563 \n", "... ... ... ... \n", "5728 no no 0.9970 \n", "25520 no no 0.9964 \n", "26232 no no 0.9913 \n", "19434 no no 0.9900 \n", "30192 no no 0.9249 \n", "\n", "[11529 rows x 19 columns]" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "predict_model(final_gbc)" ] }, { "cell_type": "code", "execution_count": 25, "id": "c1816ccd", "metadata": { "scrolled": true }, "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", "
 ModelAccuracyAUCRecallPrec.F1KappaMCC
0Gradient Boosting Classifier0.90680.93400000.50020.5059
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "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", "
agejobmaritaleducationdefaultbalancehousingloancontactdaymonthdurationcampaignpdayspreviouspoutcomedepositprediction_labelprediction_score
033entrepreneurmarriedsecondaryno2yesyesunknown5may761-10unknownnono0.9993
133unknownsingleunknownno1nonounknown5may1981-10unknownnono0.9978
245admin.singleunknownno13yesnounknown5may981-10unknownnono0.9979
350managementmarriedsecondaryno49yesnounknown5may1802-10unknownnono0.9967
432managementmarriedtertiaryno0yesnounknown5may1791-10unknownnono0.9967
\n", "
" ], "text/plain": [ " age job marital education default balance housing loan \\\n", "0 33 entrepreneur married secondary no 2 yes yes \n", "1 33 unknown single unknown no 1 no no \n", "2 45 admin. single unknown no 13 yes no \n", "3 50 management married secondary no 49 yes no \n", "4 32 management married tertiary no 0 yes no \n", "\n", " contact day month duration campaign pdays previous poutcome deposit \\\n", "0 unknown 5 may 76 1 -1 0 unknown no \n", "1 unknown 5 may 198 1 -1 0 unknown no \n", "2 unknown 5 may 98 1 -1 0 unknown no \n", "3 unknown 5 may 180 2 -1 0 unknown no \n", "4 unknown 5 may 179 1 -1 0 unknown no \n", "\n", " prediction_label prediction_score \n", "0 no 0.9993 \n", "1 no 0.9978 \n", "2 no 0.9979 \n", "3 no 0.9967 \n", "4 no 0.9967 " ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "unseen_predictions = predict_model(final_gbc, data=data_unseen)\n", "unseen_predictions.head()" ] }, { "cell_type": "markdown", "id": "1b576255", "metadata": {}, "source": [ "## Save Model " ] }, { "cell_type": "code", "execution_count": 26, "id": "c584299c", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Transformation Pipeline and Model Successfully Saved\n" ] }, { "data": { "text/plain": [ "(Pipeline(memory=FastMemory(location=C:\\Users\\owner\\AppData\\Local\\Temp\\joblib),\n", " steps=[('label_encoding',\n", " TransformerWrapperWithInverse(exclude=None, include=None,\n", " transformer=LabelEncoder())),\n", " ('numerical_imputer',\n", " TransformerWrapper(exclude=None,\n", " include=['age', 'balance', 'day',\n", " 'duration', 'campaign', 'pdays',\n", " 'previous'],\n", " transformer=SimpleImputer(add_indica...\n", " criterion='friedman_mse', init=None,\n", " learning_rate=0.3, loss='log_loss',\n", " max_depth=5, max_features='sqrt',\n", " max_leaf_nodes=None,\n", " min_impurity_decrease=0.01,\n", " min_samples_leaf=3,\n", " min_samples_split=4,\n", " min_weight_fraction_leaf=0.0,\n", " n_estimators=80,\n", " n_iter_no_change=None,\n", " random_state=321, subsample=0.95,\n", " tol=0.0001, validation_fraction=0.1,\n", " verbose=0, warm_start=False))],\n", " verbose=False),\n", " './Pycaret/Final_gbc.pkl')" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "save_model(final_gbc, './Pycaret/Final_gbc')" ] }, { "cell_type": "markdown", "id": "3ec75d79", "metadata": {}, "source": [ "## Load Model" ] }, { "cell_type": "code", "execution_count": 27, "id": "ec0008ee", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Transformation Pipeline and Model Successfully Loaded\n" ] } ], "source": [ "saved_final_gbc = load_model('./Pycaret/Final_gbc')" ] }, { "cell_type": "code", "execution_count": 28, "id": "8b1d03d7", "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", "
 ModelAccuracyAUCRecallPrec.F1KappaMCC
0Gradient Boosting Classifier0.90680.93400000.50020.5059
\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "new_prediction = predict_model(saved_final_gbc, data=data_unseen)" ] }, { "cell_type": "code", "execution_count": 29, "id": "509c9d77", "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", "
agejobmaritaleducationdefaultbalancehousingloancontactdaymonthdurationcampaignpdayspreviouspoutcomedepositprediction_labelprediction_score
033entrepreneurmarriedsecondaryno2yesyesunknown5may761-10unknownnono0.9993
133unknownsingleunknownno1nonounknown5may1981-10unknownnono0.9978
245admin.singleunknownno13yesnounknown5may981-10unknownnono0.9979
350managementmarriedsecondaryno49yesnounknown5may1802-10unknownnono0.9967
432managementmarriedtertiaryno0yesnounknown5may1791-10unknownnono0.9967
\n", "
" ], "text/plain": [ " age job marital education default balance housing loan \\\n", "0 33 entrepreneur married secondary no 2 yes yes \n", "1 33 unknown single unknown no 1 no no \n", "2 45 admin. single unknown no 13 yes no \n", "3 50 management married secondary no 49 yes no \n", "4 32 management married tertiary no 0 yes no \n", "\n", " contact day month duration campaign pdays previous poutcome deposit \\\n", "0 unknown 5 may 76 1 -1 0 unknown no \n", "1 unknown 5 may 198 1 -1 0 unknown no \n", "2 unknown 5 may 98 1 -1 0 unknown no \n", "3 unknown 5 may 180 2 -1 0 unknown no \n", "4 unknown 5 may 179 1 -1 0 unknown no \n", "\n", " prediction_label prediction_score \n", "0 no 0.9993 \n", "1 no 0.9978 \n", "2 no 0.9979 \n", "3 no 0.9967 \n", "4 no 0.9967 " ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "new_prediction.head()" ] }, { "cell_type": "code", "execution_count": null, "id": "beeb95ab", "metadata": {}, "outputs": [], "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": "3.10.9" } }, "nbformat": 4, "nbformat_minor": 5 }