{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Amazon SageMaker XGBoost를 이용한 다이렉트 마케팅 타게팅\n", "\n", "_**Gradient Boosted Trees를 이용하는 지도학습: 편향된 클래스의 이진 분류 예측문제 해결 **_\n", "\n", "---\n", "\n", "본 노트북은 다음 소스를 한글로 번역하고 일부 코드를 수정하였습니다. \n", "- https://github.com/awslabs/amazon-sagemaker-examples/blob/master/introduction_to_applying_machine_learning/xgboost_direct_marketing/xgboost_direct_marketing_sagemaker.ipynb \n", "\n", "---\n", "\n", "## 목차\n", "\n", "1. [배경](#배경)\n", "1. [준비](#준비)\n", "1. [데이터](#데이터)\n", " 1. [탐험](#탐험)\n", " 1. [변형](#변형)\n", "1. [학습](#학습)\n", "1. [호스팅](#호스팅)\n", "1. [평가](#평가)\n", "1. [확장](#확장)\n", "\n", "---\n", "\n", "## 배경\n", "메일, 이메일, 전화 등을 이용하는 다이렉트 마케팅은 고객을 모객하는 일반적인 방법입니다. 우리의 자원과 고객의 시간이 제한적이므로 우리는 특정 제안에 참여할 가능성이 높은 고객의 서브그룹에 포키싱할 필요가 있습니다. 인구통계정보나 과거 상호작용, 환경요인 등을 고려하여 이런 잠재고객을 예측하는 것은 머신러닝의 일반적인 문제입니다. \n", "\n", "본 노트북은 한 번 또는 그 이상의 전화요청으로 고객이 은행 정기예금(term deposit)에 가입할 지를 예측하는 문제 사례입니다. 진행 단계는 다음과 같습니다.\n", "\n", "* Amazon SageMaker notebook을 준비합니다.\n", "* 인터넷에서 Amazon SageMaker로 데이터를 다운로드합니다.\n", "* 데이터를 조사하고 SageMaker 알고리즘에서 사용할 수 있도록 변형합니다.\n", "* Gradient Boosting 알고리즘을 이용하여 모델을 학습합니다.\n", "* 모델의 성능을 검증합니다.\n", "* 앞으로의 예측에 모델을 적용합니다.\n", "\n", "\n", "---\n", "\n", "## 준비\n", "\n", "_본 노트북은 ml.m4.xlarge 인스턴스에서 테스트되었습니다._\n", "\n", "다음을 정의합니다.\n", "Let's start by specifying:\n", "\n", "- S3 버킷과 prefix : 노트북 인스턴스, 학습, 호스팅 인스턴스와 동일한 리전에 있어야 합니다. \n", "- IAM 역할(role) arn : 학습, 호스팅작업이 데이터에 접근할 때 사용합니다. " ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "# !pip install -U sagemaker" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'2.24.0'" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import sagemaker\n", "sagemaker.__version__" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "isConfigCell": true, "tags": [ "parameters" ] }, "outputs": [], "source": [ "import sagemaker\n", "\n", "sagemaker_session = sagemaker.Session()\n", "bucket = sagemaker.Session().default_bucket() # replace with an existing bucket if needed\n", "prefix = 'sagemaker/DEMO-xgboost-dm' # prefix used for all data stored within the bucket\n", "\n", "\n", "# Define IAM role\n", "import boto3\n", "from sagemaker import get_execution_role\n", "\n", "role = get_execution_role()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "분석과정에서 필요한 파이썬 라이브러리를 로드합니다." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "import numpy as np # For matrix operations and numerical processing\n", "import pandas as pd # For munging tabular data\n", "import matplotlib.pyplot as plt # For charts and visualizations\n", "from IPython.display import Image # For displaying images in the notebook\n", "from IPython.display import display # For displaying outputs in the notebook\n", "from time import gmtime, strftime # For labeling SageMaker models, endpoints, etc.\n", "import sys # For writing outputs to notebook\n", "import math # For ceiling function\n", "import json # For parsing hosting outputs\n", "import os # For manipulating filepath names\n", "import sagemaker # Amazon SageMaker's Python SDK provides many helper functions\n", "# from sagemaker.predictor import csv_serializer # Converts strings for HTTP POST requests on inference" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "\n", "## 데이터\n", "[direct marketing dataset](https://sagemaker-sample-data-us-west-2.s3-us-west-2.amazonaws.com/autopilot/direct_marketing/bank-additional.zip) 데이터를 다운로드합니다.\n", "\n", "\\[Moro et al., 2014\\] S. Moro, P. Cortez and P. Rita. A Data-Driven Approach to Predict the Success of Bank Telemarketing. Decision Support Systems, Elsevier, 62:22-31, June 2014\n", "\n", "\n" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "--2021-01-23 06:33:48-- https://sagemaker-sample-data-us-west-2.s3-us-west-2.amazonaws.com/autopilot/direct_marketing/bank-additional.zip\n", "Resolving sagemaker-sample-data-us-west-2.s3-us-west-2.amazonaws.com (sagemaker-sample-data-us-west-2.s3-us-west-2.amazonaws.com)... 52.218.152.105\n", "Connecting to sagemaker-sample-data-us-west-2.s3-us-west-2.amazonaws.com (sagemaker-sample-data-us-west-2.s3-us-west-2.amazonaws.com)|52.218.152.105|:443... connected.\n", "HTTP request sent, awaiting response... 200 OK\n", "Length: 432828 (423K) [application/zip]\n", "Saving to: ‘bank-additional.zip.3’\n", "\n", "bank-additional.zip 100%[===================>] 422.68K 1.16MB/s in 0.4s \n", "\n", "2021-01-23 06:33:49 (1.16 MB/s) - ‘bank-additional.zip.3’ saved [432828/432828]\n", "\n", "/bin/sh: apt-get: command not found\n", "Archive: bank-additional.zip\n", " inflating: bank-additional/bank-additional-names.txt \n", " inflating: bank-additional/bank-additional.csv \n", " inflating: bank-additional/bank-additional-full.csv \n" ] } ], "source": [ "!wget https://sagemaker-sample-data-us-west-2.s3-us-west-2.amazonaws.com/autopilot/direct_marketing/bank-additional.zip\n", "!apt-get install unzip -y\n", "!unzip -o bank-additional.zip" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\n", "pandas 데이터프레임으로 로드하고 내용을 살펴봅니다." ] }, { "cell_type": "code", "execution_count": 6, "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
agejobmaritaleducationdefaulthousingloancontactmonthday_of_weekdurationcampaignpdayspreviouspoutcomeemp.var.ratecons.price.idxcons.conf.idxeuribor3mnr.employedy
056housemaidmarriedbasic.4ynononotelephonemaymon26119990nonexistent1.193.994-36.44.8575191.0no
157servicesmarriedhigh.schoolunknownnonotelephonemaymon14919990nonexistent1.193.994-36.44.8575191.0no
237servicesmarriedhigh.schoolnoyesnotelephonemaymon22619990nonexistent1.193.994-36.44.8575191.0no
340admin.marriedbasic.6ynononotelephonemaymon15119990nonexistent1.193.994-36.44.8575191.0no
456servicesmarriedhigh.schoolnonoyestelephonemaymon30719990nonexistent1.193.994-36.44.8575191.0no
..................................................................
4118373retiredmarriedprofessional.coursenoyesnocellularnovfri33419990nonexistent-1.194.767-50.81.0284963.6yes
4118446blue-collarmarriedprofessional.coursenononocellularnovfri38319990nonexistent-1.194.767-50.81.0284963.6no
4118556retiredmarrieduniversity.degreenoyesnocellularnovfri18929990nonexistent-1.194.767-50.81.0284963.6no
4118644technicianmarriedprofessional.coursenononocellularnovfri44219990nonexistent-1.194.767-50.81.0284963.6yes
4118774retiredmarriedprofessional.coursenoyesnocellularnovfri23939991failure-1.194.767-50.81.0284963.6no
\n", "

41188 rows × 21 columns

\n", "
" ], "text/plain": [ " age job marital education default housing loan \\\n", "0 56 housemaid married basic.4y no no no \n", "1 57 services married high.school unknown no no \n", "2 37 services married high.school no yes no \n", "3 40 admin. married basic.6y no no no \n", "4 56 services married high.school no no yes \n", "... ... ... ... ... ... ... ... \n", "41183 73 retired married professional.course no yes no \n", "41184 46 blue-collar married professional.course no no no \n", "41185 56 retired married university.degree no yes no \n", "41186 44 technician married professional.course no no no \n", "41187 74 retired married professional.course no yes no \n", "\n", " contact month day_of_week duration campaign pdays previous \\\n", "0 telephone may mon 261 1 999 0 \n", "1 telephone may mon 149 1 999 0 \n", "2 telephone may mon 226 1 999 0 \n", "3 telephone may mon 151 1 999 0 \n", "4 telephone may mon 307 1 999 0 \n", "... ... ... ... ... ... ... ... \n", "41183 cellular nov fri 334 1 999 0 \n", "41184 cellular nov fri 383 1 999 0 \n", "41185 cellular nov fri 189 2 999 0 \n", "41186 cellular nov fri 442 1 999 0 \n", "41187 cellular nov fri 239 3 999 1 \n", "\n", " poutcome emp.var.rate cons.price.idx cons.conf.idx euribor3m \\\n", "0 nonexistent 1.1 93.994 -36.4 4.857 \n", "1 nonexistent 1.1 93.994 -36.4 4.857 \n", "2 nonexistent 1.1 93.994 -36.4 4.857 \n", "3 nonexistent 1.1 93.994 -36.4 4.857 \n", "4 nonexistent 1.1 93.994 -36.4 4.857 \n", "... ... ... ... ... ... \n", "41183 nonexistent -1.1 94.767 -50.8 1.028 \n", "41184 nonexistent -1.1 94.767 -50.8 1.028 \n", "41185 nonexistent -1.1 94.767 -50.8 1.028 \n", "41186 nonexistent -1.1 94.767 -50.8 1.028 \n", "41187 failure -1.1 94.767 -50.8 1.028 \n", "\n", " nr.employed y \n", "0 5191.0 no \n", "1 5191.0 no \n", "2 5191.0 no \n", "3 5191.0 no \n", "4 5191.0 no \n", "... ... ... \n", "41183 4963.6 yes \n", "41184 4963.6 no \n", "41185 4963.6 no \n", "41186 4963.6 yes \n", "41187 4963.6 no \n", "\n", "[41188 rows x 21 columns]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data = pd.read_csv('./bank-additional/bank-additional-full.csv')\n", "pd.set_option('display.max_columns', 500) # Make sure we can see all of the columns\n", "pd.set_option('display.max_rows', 20) # Keep the output on one page\n", "data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "데이터를 살펴봅니다. 대략적으로 다음 내용이 눈에 띕니다.\n", "Let's talk about the data. At a high level, we can see:\n", "\n", "* 3만건 이상의 고객 레코드가 있고 각 고객에 대하여 20건의 속성값(feature)이 습니다.\n", "* 속성은 숫자 또는 명목형(categorical) 값이 섞여 있습니다.\n", "* 데이터는 `time`과 `contact` 등에 따라 정렬된 것처럼 보입니다. \n", "\n", "_**속성 정보:**_\n", "\n", "*인구통계정보:*\n", "* `age`: 고객의 나이 (numeric)\n", "* `job`: 직업군 (categorical: 'admin.', 'services', ...)\n", "* `marital`: 결혼여부 (categorical: 'married', 'single', ...)\n", "* `education`: 학업 (categorical: 'basic.4y', 'high.school', ...)\n", "\n", "*과거 고객 이벤트:*\n", "* `default`: dafault 여부? (categorical: 'no', 'unknown', ...)\n", "* `housing`: 주택대출 여부? (categorical: 'no', 'yes', ...)\n", "* `loan`: 개인대출 여부? (categorical: 'no', 'yes', ...)\n", "\n", "*과거 다이렉트 마케팅 이력:*\n", "* `contact`: 커뮤니케이션 유형 (categorical: 'cellular', 'telephone', ...)\n", "* `month`: 최종 접촉 월 (categorical: 'may', 'nov', ...)\n", "* `day_of_week`: 최종 접촉 주 (categorical: 'mon', 'fri', ...)\n", "* `duration`: 최종 접촉 기간(초) (numeric). 중요: duration 이 0 이면 `y`는 'no'임.\n", " \n", "*캠페인 정보:*\n", "* `campaign`: 이번 캠페인 동안 접촉한 회수 (numeric, includes last contact)\n", "* `pdays`: 이전 캠페인에서 접촉한 마지막 날짜 이후 경과 시간(일) (numeric)\n", "* `previous`: 이번 캠페인 이전에 고객과의 접촉 회수 (numeric)\n", "* `poutcome`: 이전 캠페인의 성과 (categorical: 'nonexistent','success', ...)\n", "\n", "*외부 환경 요인:*\n", "* `emp.var.rate`: 고용 변화율(Employment variation rate, quarterly) (numeric)\n", "* `cons.price.idx`: 소비자 가격 지수(Consumer price index, monthly) (numeric)\n", "* `cons.conf.idx`: 고객 확신 지수 (Consumer confidence index, monthly) (numeric)\n", "* `euribor3m`: Euribor 3 개월 비율 (daily) (numeric)\n", "* `nr.employed`: 고용자 수(Number of employees, quarterly) (numeric)\n", "\n", "*타겟 변수:*\n", "* `y`: 고객이 정기예금에 가입하였는가? (binary: 'yes','no')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 탐험\n", "데이터 탐험(EDA)를 해봅니다. 첫번째로 속성의 분포를 살펴봅니다. " ] }, { "cell_type": "code", "execution_count": 7, "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", "
col_0% observations
job
admin.0.253035
blue-collar0.224677
entrepreneur0.035350
housemaid0.025736
management0.070992
retired0.041760
self-employed0.034500
services0.096363
student0.021244
technician0.163713
unemployed0.024619
unknown0.008012
\n", "
" ], "text/plain": [ "col_0 % observations\n", "job \n", "admin. 0.253035\n", "blue-collar 0.224677\n", "entrepreneur 0.035350\n", "housemaid 0.025736\n", "management 0.070992\n", "retired 0.041760\n", "self-employed 0.034500\n", "services 0.096363\n", "student 0.021244\n", "technician 0.163713\n", "unemployed 0.024619\n", "unknown 0.008012" ] }, "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", "
col_0% observations
marital
divorced0.111974
married0.605225
single0.280859
unknown0.001942
\n", "
" ], "text/plain": [ "col_0 % observations\n", "marital \n", "divorced 0.111974\n", "married 0.605225\n", "single 0.280859\n", "unknown 0.001942" ] }, "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", "
col_0% observations
education
basic.4y0.101389
basic.6y0.055647
basic.9y0.146766
high.school0.231014
illiterate0.000437
professional.course0.127294
university.degree0.295426
unknown0.042027
\n", "
" ], "text/plain": [ "col_0 % observations\n", "education \n", "basic.4y 0.101389\n", "basic.6y 0.055647\n", "basic.9y 0.146766\n", "high.school 0.231014\n", "illiterate 0.000437\n", "professional.course 0.127294\n", "university.degree 0.295426\n", "unknown 0.042027" ] }, "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", "
col_0% observations
default
no0.791201
unknown0.208726
yes0.000073
\n", "
" ], "text/plain": [ "col_0 % observations\n", "default \n", "no 0.791201\n", "unknown 0.208726\n", "yes 0.000073" ] }, "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", "
col_0% observations
housing
no0.452122
unknown0.024036
yes0.523842
\n", "
" ], "text/plain": [ "col_0 % observations\n", "housing \n", "no 0.452122\n", "unknown 0.024036\n", "yes 0.523842" ] }, "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", "
col_0% observations
loan
no0.824269
unknown0.024036
yes0.151695
\n", "
" ], "text/plain": [ "col_0 % observations\n", "loan \n", "no 0.824269\n", "unknown 0.024036\n", "yes 0.151695" ] }, "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", "
col_0% observations
contact
cellular0.634748
telephone0.365252
\n", "
" ], "text/plain": [ "col_0 % observations\n", "contact \n", "cellular 0.634748\n", "telephone 0.365252" ] }, "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", "
col_0% observations
month
apr0.063902
aug0.149995
dec0.004419
jul0.174177
jun0.129115
mar0.013256
may0.334296
nov0.099568
oct0.017432
sep0.013839
\n", "
" ], "text/plain": [ "col_0 % observations\n", "month \n", "apr 0.063902\n", "aug 0.149995\n", "dec 0.004419\n", "jul 0.174177\n", "jun 0.129115\n", "mar 0.013256\n", "may 0.334296\n", "nov 0.099568\n", "oct 0.017432\n", "sep 0.013839" ] }, "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", "
col_0% observations
day_of_week
fri0.190031
mon0.206711
thu0.209357
tue0.196416
wed0.197485
\n", "
" ], "text/plain": [ "col_0 % observations\n", "day_of_week \n", "fri 0.190031\n", "mon 0.206711\n", "thu 0.209357\n", "tue 0.196416\n", "wed 0.197485" ] }, "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", "
col_0% observations
poutcome
failure0.103234
nonexistent0.863431
success0.033335
\n", "
" ], "text/plain": [ "col_0 % observations\n", "poutcome \n", "failure 0.103234\n", "nonexistent 0.863431\n", "success 0.033335" ] }, "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", "
col_0% observations
y
no0.887346
yes0.112654
\n", "
" ], "text/plain": [ "col_0 % observations\n", "y \n", "no 0.887346\n", "yes 0.112654" ] }, "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", "
agedurationcampaignpdayspreviousemp.var.ratecons.price.idxcons.conf.idxeuribor3mnr.employed
count41188.0000041188.00000041188.00000041188.00000041188.00000041188.00000041188.00000041188.00000041188.00000041188.000000
mean40.02406258.2850102.567593962.4754540.1729630.08188693.575664-40.5026003.6212915167.035911
std10.42125259.2792492.770014186.9109070.4949011.5709600.5788404.6281981.73444772.251528
min17.000000.0000001.0000000.0000000.000000-3.40000092.201000-50.8000000.6340004963.600000
25%32.00000102.0000001.000000999.0000000.000000-1.80000093.075000-42.7000001.3440005099.100000
50%38.00000180.0000002.000000999.0000000.0000001.10000093.749000-41.8000004.8570005191.000000
75%47.00000319.0000003.000000999.0000000.0000001.40000093.994000-36.4000004.9610005228.100000
max98.000004918.00000056.000000999.0000007.0000001.40000094.767000-26.9000005.0450005228.100000
\n", "
" ], "text/plain": [ " age duration campaign pdays previous \\\n", "count 41188.00000 41188.000000 41188.000000 41188.000000 41188.000000 \n", "mean 40.02406 258.285010 2.567593 962.475454 0.172963 \n", "std 10.42125 259.279249 2.770014 186.910907 0.494901 \n", "min 17.00000 0.000000 1.000000 0.000000 0.000000 \n", "25% 32.00000 102.000000 1.000000 999.000000 0.000000 \n", "50% 38.00000 180.000000 2.000000 999.000000 0.000000 \n", "75% 47.00000 319.000000 3.000000 999.000000 0.000000 \n", "max 98.00000 4918.000000 56.000000 999.000000 7.000000 \n", "\n", " emp.var.rate cons.price.idx cons.conf.idx euribor3m nr.employed \n", "count 41188.000000 41188.000000 41188.000000 41188.000000 41188.000000 \n", "mean 0.081886 93.575664 -40.502600 3.621291 5167.035911 \n", "std 1.570960 0.578840 4.628198 1.734447 72.251528 \n", "min -3.400000 92.201000 -50.800000 0.634000 4963.600000 \n", "25% -1.800000 93.075000 -42.700000 1.344000 5099.100000 \n", "50% 1.100000 93.749000 -41.800000 4.857000 5191.000000 \n", "75% 1.400000 93.994000 -36.400000 4.961000 5228.100000 \n", "max 1.400000 94.767000 -26.900000 5.045000 5228.100000 " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Frequency tables for each categorical feature\n", "for column in data.select_dtypes(include=['object']).columns:\n", " display(pd.crosstab(index=data[column], columns='% observations', normalize='columns'))\n", "\n", "# Histograms for each numeric features\n", "display(data.describe())\n", "%matplotlib inline\n", "hist = data.hist(bins=30, sharey=True, figsize=(10, 10))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "다음을 생각해 볼 수 있습니다.:\n", "\n", "* `y`값의 약 90%가 \"no\"입니다. 곧, 대부분의 고객은 정기예금에 가입하지 않았습니다.\n", "* 많은 속성값이 \"unknown\"상태이고 그 분포는 속성에 따라 다릅니다. \"unknown\"상태를 다룰 때 해당 값의 원인이 무엇일 지 어떻게 다루어야 할 지 주의가 필요합니다.\n", " * \"unknown\"이 고유한 카테고리를 가지고 있지만 실제로는 발생 원인에 따라 다른 카테고리중 하나에 포함될 것입니다.\n", "* 속성들이 소수의 관찰값을 보유한 카테고리를 가지고 있는 경우가 많습니다. 만약 어떤 작은 카테고리의 속성이 타겟값에 높은 영향을 미친다면, 이를 일반화할 수 있을 정도로 충분히 많은 수의 데이터 샘플(evidence)을 가지고 있나요?\n", "* 접촉 타이밍 정보가 과도하게 치우쳐(skewed) 있습니다. 약 1/3이 5월에 이루어졌으며 12월은 1%미만입니다. 우리가 만약 다음 12월에 대한 예측을 실행한다면 이것이 어떤 의미를 가질까요?\n", "* 숫자형 데이터에 결측치는 없습니다. 결측값은 이미 보정된 상태입니다.\n", " * `pdays`는 대부분의 고객들에 대하여 1000이 넘습니다. 이전 컨택이 많지 않았음을 알려줍니다.\n", "* 일부 숫자형 속성은 긴 롱테일을 가지고 있습니다. 속성의 주요 부분과 롱테일 부분을 분리해서 접근해야 할 필요가 있나요?\n", "* 일부 숫자형 속성(특히 외부 시장지표 값들)은 몇가지 그룹으로 나누어질 수 있어 보입니다. 이들을 명목형 변수로 바꾸어야 할까요?\n", "\n", "다음은, 이들 속성들이 예측하고자 하는 타겟값과 어떻게 관련되는지 살펴봅니다." ] }, { "cell_type": "code", "execution_count": 8, "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", "
ynoyes
job
admin.0.2481670.291379
blue-collar0.2357450.137500
entrepreneur0.0364450.026724
housemaid0.0261030.022845
management0.0710300.070690
retired0.0351870.093534
self-employed0.0348040.032112
services0.0997590.069612
student0.0164170.059267
technician0.1645230.157328
unemployed0.0238040.031034
unknown0.0080170.007974
\n", "
" ], "text/plain": [ "y no yes\n", "job \n", "admin. 0.248167 0.291379\n", "blue-collar 0.235745 0.137500\n", "entrepreneur 0.036445 0.026724\n", "housemaid 0.026103 0.022845\n", "management 0.071030 0.070690\n", "retired 0.035187 0.093534\n", "self-employed 0.034804 0.032112\n", "services 0.099759 0.069612\n", "student 0.016417 0.059267\n", "technician 0.164523 0.157328\n", "unemployed 0.023804 0.031034\n", "unknown 0.008017 0.007974" ] }, "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", "
ynoyes
marital
divorced0.1131660.102586
married0.6127830.545690
single0.2721900.349138
unknown0.0018610.002586
\n", "
" ], "text/plain": [ "y no yes\n", "marital \n", "divorced 0.113166 0.102586\n", "married 0.612783 0.545690\n", "single 0.272190 0.349138\n", "unknown 0.001861 0.002586" ] }, "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", "
ynoyes
education
basic.4y0.1025500.092241
basic.6y0.0575680.040517
basic.9y0.1524570.101940
high.school0.2321330.222198
illiterate0.0003830.000862
professional.course0.1271750.128233
university.degree0.2872390.359914
unknown0.0404950.054095
\n", "
" ], "text/plain": [ "y no yes\n", "education \n", "basic.4y 0.102550 0.092241\n", "basic.6y 0.057568 0.040517\n", "basic.9y 0.152457 0.101940\n", "high.school 0.232133 0.222198\n", "illiterate 0.000383 0.000862\n", "professional.course 0.127175 0.128233\n", "university.degree 0.287239 0.359914\n", "unknown 0.040495 0.054095" ] }, "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", "
ynoyes
default
no0.7768140.904526
unknown0.2231040.095474
yes0.0000820.000000
\n", "
" ], "text/plain": [ "y no yes\n", "default \n", "no 0.776814 0.904526\n", "unknown 0.223104 0.095474\n", "yes 0.000082 0.000000" ] }, "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", "
ynoyes
housing
no0.4540880.436638
unknown0.0241600.023060
yes0.5217520.540302
\n", "
" ], "text/plain": [ "y no yes\n", "housing \n", "no 0.454088 0.436638\n", "unknown 0.024160 0.023060\n", "yes 0.521752 0.540302" ] }, "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", "
ynoyes
loan
no0.8235740.829741
unknown0.0241600.023060
yes0.1522660.147198
\n", "
" ], "text/plain": [ "y no yes\n", "loan \n", "no 0.823574 0.829741\n", "unknown 0.024160 0.023060\n", "yes 0.152266 0.147198" ] }, "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", "
ynoyes
contact
cellular0.609910.830388
telephone0.390090.169612
\n", "
" ], "text/plain": [ "y no yes\n", "contact \n", "cellular 0.60991 0.830388\n", "telephone 0.39009 0.169612" ] }, "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", "
ynoyes
month
apr0.0572670.116164
aug0.1511160.141164
dec0.0025450.019181
jul0.1785320.139871
jun0.1302120.120474
mar0.0073880.059483
may0.3524950.190948
nov0.1008260.089655
oct0.0110270.067888
sep0.0085910.055172
\n", "
" ], "text/plain": [ "y no yes\n", "month \n", "apr 0.057267 0.116164\n", "aug 0.151116 0.141164\n", "dec 0.002545 0.019181\n", "jul 0.178532 0.139871\n", "jun 0.130212 0.120474\n", "mar 0.007388 0.059483\n", "may 0.352495 0.190948\n", "nov 0.100826 0.089655\n", "oct 0.011027 0.067888\n", "sep 0.008591 0.055172" ] }, "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", "
ynoyes
day_of_week
fri0.1910090.182328
mon0.2097790.182543
thu0.2073440.225216
tue0.1952770.205388
wed0.1965910.204526
\n", "
" ], "text/plain": [ "y no yes\n", "day_of_week \n", "fri 0.191009 0.182328\n", "mon 0.209779 0.182543\n", "thu 0.207344 0.225216\n", "tue 0.195277 0.205388\n", "wed 0.196591 0.204526" ] }, "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", "
ynoyes
poutcome
failure0.0997870.130388
nonexistent0.8871070.676940
success0.0131060.192672
\n", "
" ], "text/plain": [ "y no yes\n", "poutcome \n", "failure 0.099787 0.130388\n", "nonexistent 0.887107 0.676940\n", "success 0.013106 0.192672" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "age\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "duration\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "campaign\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "pdays\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "previous\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "emp.var.rate\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "cons.price.idx\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEKCAYAAAAFJbKyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAcOUlEQVR4nO3dfbBdVZ3m8e9jkFdBkhAQCDFRMnYrrYBpwHaqhyECQSzjVEE3jC2Rjp2qbnxr2tbQPdPMqNihxmkUx45DQ1ri0AQm6pCRCKZ5GcoqCYQXebdJQ4RIJMGE+EKJBp/5Y69LTm7OvTf7nnteeT5Vt87Za6997m/D2fndtfbaa8k2ERERe+pV3Q4gIiL6SxJHRETUksQRERG1JHFEREQtSRwREVFLEkdERNSSxBEREbUkcURERC1JHBERUUsSxwCRtEHSJyQ9IGm7pOsk7Vv2/Ymk9ZK2Slol6YhuxxsxUST9paSvDyv7kqQvSHqtpKskbZL0I0mflTSp1Dla0v8r18tzkq7rzhn0lySOwfMHwDxgFvBW4IOSTgH+tuw7HPghsKJrEUZMvP8FzJN0MICkvYA/BL4GXA3sAI4GjgNOAz5UjvsM8B1gMjAd+FJnw+5PSRyD53Lbz9jeCvxf4Fjg/cAy2/fafhG4CHiHpJndCzNi4tjeBNwBnF2K5gHPARuBM4CP2/6F7c3AZcA5pd6vgdcDR9j+pe3vdjby/pTEMXh+3PD+BeA1wBFUrQwAbP8c+AlwZGdDi2irq4E/Ku//iKq18Xrg1cAmSc9Leh74n8Chpd4nAQF3SXpY0h93OOa+tFe3A4iOeIbqAgJA0gHAVOBHXYsoYuL9H2CppGOA91AlhV8DLwKH2N4x/ADbPwb+BEDSvwX+WdIdttd3Luz+kxbHK8M/AedLOlbSPsDngLW2N3Q3rIiJY/uXwEqq7/tdtp8qXVjfAf67pIMkvUrSGyX9OwBJZ0uaXj5iG2DgpW7E30+SOF4BbN8C/Gfg68Am4I3s7OONGCRXA79D1U015Dxgb+ARquSwkmqQCMDvAmsl/RxYBXzM9pOdC7c/KQs5RcSgkDQDeAx4ne2fdjueQZUWR0QMBEmvAi4EViRptFdujkdE3ysDPp6lGj04r8vhDLx0VUVERC3pqoqIiFqSOCIiopa+vcdxyCGHeObMmd0OIwbMPffc85ztad2Oo45cC9EOo10LfZs4Zs6cybp167odRgwYST8cu1ZvybUQ7TDatZCuqoiIqCWJI6LNJP15mUDvIUnXStpX0ixJayU9XtZN2bvU3adsry/7Z3Y3+ojdjZk4JC2TtFnSQ032fUKSJR1StiXp8vKlf0DS8Q11F5SL5HFJCxrK3y7pwXLM5ZI0UScX0W2SjgQ+CsyxfQwwiWq6l0uBy2zPppoGY2E5ZCGwzfbRVNN/X9r5qCNGtyctjq/S5IEaSUcBpwJPNRSfAcwuP4uApaXuFOBi4ETgBOBiSZPLMUtL3aHj8vBODJq9gP3K4kL7U80XdgrVnElQza/0vvJ+ftmm7J+bP6ai14yZOGzfAWxtsusyqmmLG58gnA8sd+VO4GBJhwOnA2tsb7W9DVhDtVrX4cBBtr/n6knE5ey8gCL6nu0fAZ+n+gNrE7AduAd4vmGa743sXBvlSODpcuyOUn9qJ2OOGMu47nFIei/wI9vfH7br5S99MXRBjFa+sUn5SL93kaR1ktZt2bJlPKFHdFRpWc+nWsr3COAAqpb5cEN/gDVrXew2vUOuheim2olD0v7AXwN/02x3kzKPo7wp21fYnmN7zrRpfTXUPl653gU8aXuL7V8D3wB+j6o1PjQcfjrVYltQ/fF0FLy8bvZradLiz7UQ3TSeFscbqf56+r6kDVRf+nslvY6GL30xdEGMVj69SXnEoHgKOEnS/uVexVyqdSFuA84qdRYAN5T3q8o2Zf+tzoRy0WNqPwBo+0F2rtdLSR5zbD8naRXwYUkrqG6Eb7e9SdLNwOcaboifBlxke6ukn0k6CVhLteDKl1o7peikmYtv3GV7w5IzuxRJb7K9VtJK4F5gB3AfcAVwI7BC0mdL2VXlkKuAr0laT9XSGKgFt4Z/XyDfmX40ZuKQdC1wMnCIpI3AxbavGqH6auDdwHrgBeB8gJIgPgPcXep92vZQ8/tPqUZu7Qd8u/xEDAzbF1ONKmz0BNUIw+F1fwmc3Ym4IsZrzMRh+9wx9s9seG/gghHqLQOWNSlfBxwzVhwREdEb8uR4RETUksQRERG1JHFEREQtSRwREVFLEkdERNSSxBEREbUkcURERC1JHBERUUsSR0RE1JLEERERtSRxRERELUkcERFRSxJHRETUksQRERG1JHFEREQtSRwREVFLEkdEG0l6k6T7G35+KunjkqZIWiPp8fI6udSXpMslrZf0gKTju30OEcMlcUS0ke0f2D7W9rHA26mWVP4msBi4xfZs4JayDXAGMLv8LAKWdj7qiNElcUR0zlzgX23/EJgPXF3KrwbeV97PB5a7cidwsKTDOx9qxMiSOCI65xzg2vL+MNubAMrroaX8SODphmM2lrKInjFm4pC0TNJmSQ81lP03SY+VPthvSjq4Yd9FpX/2B5JObyifV8rWS1rcUD5L0trS13udpL0n8gQjekH5Xr8X+N9jVW1S5iaft0jSOknrtmzZMhEhRuyxPWlxfBWYN6xsDXCM7bcC/wJcBCDpzVR/Vb2lHPP3kiZJmgR8mar/9s3AuaUuwKXAZaWvdxuwsKUziuhNZwD32n62bD871AVVXjeX8o3AUQ3HTQeeGf5htq+wPcf2nGnTprUx7IjdjZk4bN8BbB1W9h3bO8rmnVRfbqj6Z1fYftH2k8B64ITys972E7Z/BawA5ksScAqwshzf2NcbMUjOZWc3FcAqYEF5vwC4oaH8vDK66iRg+1CXVkSvmIh7HH8MfLu8H6l/dqTyqcDzDUko/bkxcCTtD5wKfKOheAlwqqTHy74lpXw18ATVH13/APxZB0ON2CN7tXKwpL8GdgDXDBU1qWaaJyiPUn+k37eIaogiM2bMqBVrRLfYfoHqj6TGsp9QjbIaXtfABR0KLWJcxt3ikLQAeA/w/vJlh5H7Z0cqf45quOFew8qbSr9uRET3jStxSJoHfAp4b/lrasgq4BxJ+0iaRfUQ013A3cDsMoJqb6ob6KtKwrkNOKsc39jXGxERPWhPhuNeC3wPeJOkjZIWAv8DOBBYU6ZR+AqA7YeB64FHgJuAC2y/VO5hfBi4GXgUuL7UhSoBXShpPVVz/qoJPcOIiJhQY97jsH1uk+IR/3G3fQlwSZPy1VQ3/oaXP0E16ioiIvpAnhyPiIhakjgiIqKWJI6IiKgliSMiImpJ4oiIiFpaenI8IiIqMxffuMv2hiVndimS9kuLIyIiakniiIiIWpI4IiKiliSOiIioJYkjIiJqSeKIiIhakjgiIqKWJI6IiKgliSOizSQdLGmlpMckPSrpHZKmSFoj6fHyOrnUlaTLJa2X9ICk47sdf8RwSRwR7fdF4CbbvwW8jWoxs8XALbZnA7eUbYAzqFbOnA0sApZ2PtyI0SVxRLSRpIOA36csfmb7V7afB+YDV5dqVwPvK+/nA8tduRM4WNLhHQ47YlRJHBHt9QZgC/CPku6TdKWkA4DDbG8CKK+HlvpHAk83HL+xlEX0jCSOiPbaCzgeWGr7OOAX7OyWakZNyrxbJWmRpHWS1m3ZsmViIo3YQ0kcEe21Edhoe23ZXkmVSJ4d6oIqr5sb6h/VcPx04JnhH2r7CttzbM+ZNm1a24KPaCaJI6KNbP8YeFrSm0rRXOARYBWwoJQtAG4o71cB55XRVScB24e6tCJ6xZiJQ9IySZslPdRQVnsooaQFpf7jkhY0lL9d0oPlmMslNWuqR/SzjwDXSHoAOBb4HLAEOFXS48CpZRtgNfAEsB74B+DPOh9uxOj2pMXxVWDesLJaQwklTQEuBk4ETgAuHko2pc6ihuOG/66Ivmb7/tKt9Fbb77O9zfZPbM+1Pbu8bi11bfsC22+0/Tu213U7/ojhxkwctu8Atg4rrjuU8HRgje2ttrcBa4B5Zd9Btr9n28Dyhs+KiIgeNN57HHWHEo5WvrFJeURE9KiJvjk+0lDCuuXNPzxDECMium68iaPuUMLRyqc3KW8qQxAjIrpvvImj7lDCm4HTJE0uN8VPA24u+34m6aQymuq8hs+KiIgetNdYFSRdC5wMHCJpI9XoqCXA9ZIWAk8BZ5fqq4F3Uw0lfAE4H8D2VkmfAe4u9T49NIoE+FOqkVv7Ad8uPxER0aPGTBy2zx1h19wmdQ1cMMLnLAOWNSlfBxwzVhwREdEb8uR4RETUksQRERG1JHFEREQtSRwREVFLEkdERNSSxBEREbUkcURERC1JHBERUUsSR0RE1JLEERERtSRxRLSZpA1leeT7Ja0rZbWXX47oFUkcEZ3x720fa3tO2a61/HJEL0niiOiOussvR/SMJI6I9jPwHUn3SFpUyuouvxzRM8acVj0iWvZO289IOhRYI+mxUeru0XLKJQEtApgxY8bERBmxh9LiiGgz28+U183AN4ETqL/88vDPzDLK0TVJHBFtJOkASQcOvadaNvkh6i+/HNEz0lUV0V6HAd+UBNX19k+2b5J0NzWWX47oJUkcEW1k+wngbU3Kf0LN5ZcjekW6qiIiopYkjoiIqKWlxCHpzyU9LOkhSddK2lfSLElry1QK10nau9Tdp2yvL/tnNnzORaX8B5JOb+2UIiKincadOCQdCXwUmGP7GGAScA5wKXBZmUphG7CwHLIQ2Gb7aOCyUg9Jby7HvQWYB/y9pEnjjSsiItqr1a6qvYD9JO0F7A9sAk4BVpb9w6dSGJpiYSUwV9VQk/nACtsv2n6SajTJCS3GFRERbTLuxGH7R8DnqYYSbgK2A/cAz9veUao1Tpfw8lQKZf92YCqZYiEioq+00lU1maq1MAs4AjiAambP4YamSxhpKoU9mmKh/M5FktZJWrdly5b6QUdERMta6ap6F/Ck7S22fw18A/g9qtk8h54PaZwu4eWpFMr+1wJb2cMpFiDTLERE9IJWEsdTwEmS9i/3KuYCjwC3AWeVOsOnUhiaYuEs4NbysNMq4Jwy6moW1ToEd7UQV0REtNG4nxy3vVbSSuBeYAdwH3AFcCOwQtJnS9lV5ZCrgK9JWk/V0jinfM7Dkq6nSjo7gAtsvzTeuCIior1amnLE9sXAxcOKn6DJqCjbv2TnfDzD910CXNJKLBExGGYuvnG3sg1LzuxCJDGSPDkeERG1JHFEREQtSRwREVFLEkdERNSSxBEREbUkcURERC1JHBEdIGmSpPskfats115+IKJXJHFEdMbHgEcbtmstPxDRS5I4ItpM0nTgTODKsi3qLz8Q0TOSOCLa7wvAJ4HflO2p1F9+IKJnJHFEtJGk9wCbbd/TWNyk6ljLDwz/3CwxEF2TxBHRXu8E3itpA7CCqovqC9RffmAXWWIguimJI6KNbF9ke7rtmVQzQt9q+/3UX34gomckcUR0x6eAC8syA1PZdfmBqaX8QmBxl+KLGFFL06pHxJ6zfTtwe3lfe/mBiF6RFkdERNSSxBEREbUkcURERC1JHBERUUsSR0RE1JLEERERtbSUOCQdLGmlpMckPSrpHZKmSFpTpoteI2lyqStJl5fpoh+QdHzD5ywo9R+XtGDk3xgREd3Waovji8BNtn8LeBvVtNGLgVvKdNG3sPMBpjOA2eVnEbAUQNIU4GLgRKpx7RcPJZuIiOg9404ckg4Cfp/yxKvtX9l+nl2nhR4+XfRyV+6kmqvncOB0YI3trba3AWuAeeONKyIi2quVFscbgC3AP5aVza6UdABwmO1NAOX10FL/5emii6GppEcqj4iIHtRK4tgLOB5Yavs44BeMPq/OSNNF79E00pCppCMiekEriWMjsNH22rK9kiqRPFu6oCivmxvqH9Vw/NBU0iOV7yZTSUdEdN+4E4ftHwNPS3pTKZoLPMKu00IPny76vDK66iRge+nKuhk4TdLkclP8tFIWERE9qNXZcT8CXCNpb+AJ4HyqZHS9pIXAU+yc6XM18G5gPfBCqYvtrZI+A9xd6n3a9m4L10RERG9oKXHYvh+Y02TX3CZ1DVwwwucsA5a1Eksvm7n4xt3KNiw5swuRRESvGf7vQz/825AnxyMiopYkjoiIqCWJI6KNJO0r6S5J35f0sKT/WspnSVpbptm5rtwnRNI+ZXt92T+zm/FHNJPEEdFeLwKn2H4bcCwwr4wqvBS4rEzNsw1YWOovBLbZPhq4rNSL6ClJHBFtVKbY+XnZfHX5MXAK1bNPsPvUPENT9qwE5kpq9pBsRNckcUS0maRJku6nehh2DfCvwPO2d5QqjdPsvDwFT9m/HZja2YgjRpfEEdFmtl+yfSzVrAgnAL/drFp53aMpeDL9TnRTEkdEh5TZo28HTqKaHXroOarGaXZenoKn7H8tsNsDsZl+J7opiSOijSRNk3Rweb8f8C6qdWtuA84q1YZPzTM0Zc9ZwK3l4dmIntHqlCMRMbrDgaslTaJMx2P7W5IeAVZI+ixwH2Vdm/L6NUnrqVoa53Qj6IjRJHFEtJHtB4DjmpQ/QXW/Y3j5L9k5v1tHZEqcqCuJY8DkH4GIaLfc44iIiFrS4oiIgdWPM8/2g7Q4IiKilrQ4IgZY/uKOdkiLIyIiakmLIyLGJa2ZV660OCIiopYkjoiIqCWJIyIiamk5cZS1Bu6T9K2yXXtJTEkXlfIfSDq91ZgiIqJ9JuLm+MeoZvs8qGwPLYm5QtJXqJbCXErDkpiSzin1/lDSm6kmcnsLcATwz5L+je2XJiC2GCC5GRvRG1pqcUiaDpwJXFm2Rf0lMecDK2y/aPtJYD1NJn+LiIje0GpX1ReATwK/KdtTqb8k5svlTY6JiIgeM+6uKknvATbbvkfSyUPFTaqOtSTmHi2VWX7nImARwIwZM2rFGxHRSYM8U3UrLY53Au+VtAFYQdVF9QXqL4n5cnmTY3aR5TIjIrpv3InD9kW2p9ueSXVz+1bb76f+kpirgHPKqKtZwGzgrvHGFRER7dWO5zg+BVxYlr6cyq5LYk4t5RcCiwFsPwxcDzwC3ARckBFVMSgkHSXpNkmPSnpY0sdK+RRJa8qw9TWSJpdySbq8DE9/QNLx3T2DiN1NyFxVtm8Hbi/vay+JafsS4JKJiCWix+wA/sL2vZIOBO6RtAb4IHCL7SWSFlP9IfUp4AyqVvds4ESqoewndiXyiBHkyfGINrK9yfa95f3PqJ55OpJdh6cPH7a+3JU7qe4ZHt7hsCNGlcQR0SFltoTjgLXAYbY3QZVcgENLtQxPj56XxBHRAZJeA3wd+Ljtn45WtUnZbsPTJS2StE7Sui1btkxUmBF7JOtx9LBMsTEYJL2aKmlcY/sbpfhZSYfb3lS6ojaX8j0anm77CuAKgDlz5jR97imiXdLiiGijMq3OVcCjtv+uYVfj8PThw9bPK6OrTgK2D3VpRfSKtDgi2uudwAeAByXdX8r+ClgCXC9pIfAUO0ccrgbeTTVn2wvA+Z0NtzcN8lPY49Ht/x5JHBFtZPu7NL9vATC3SX0DF7Q1qIgWpasqIiJqSeKIiIhakjgiIqKWJI6IiKgliSMiImpJ4oiIiFqSOCIiopYkjoiIqCUPAEbEQGj2NHW0R1ocERFRSxJHRETUkq6qiIgxZImDXaXFERERtYw7cUg6StJtkh6V9LCkj5XyKZLWSHq8vE4u5ZJ0uaT1kh6QdHzDZy0o9R+XtGCk3xkREd3XSlfVDuAvbN8r6UDgHklrgA8Ct9heImkxsBj4FHAGMLv8nAgsBU6UNAW4GJhDtUTmPZJW2d7WQmwR0QMy0mkwjTtxlFXJNpX3P5P0KHAkMB84uVS7GridKnHMB5aX9QbulHRwWTLzZGCN7a0AJfnMA64db2wREe30Sk+IE3KPQ9JM4DhgLXDY0FKX5fXQUu1I4OmGwzaWspHKIyKiB7U8qkrSa4CvAx+3/dNqieXmVZuUeZTyZr9rEbAIYMaMGfWD7WGv9L9gBpWkZcB7gM22jyllU4DrgJnABuAPbG8r65N/kWrp2BeAD9q+txtxR4ympRaHpFdTJY1rbH+jFD9buqAor5tL+UbgqIbDpwPPjFK+G9tX2J5je860adNaCT2iU75K1fXaaDHVfcDZwC1lG3a9D7iI6j5gRM9pZVSVgKuAR23/XcOuVcDQyKgFwA0N5eeV0VUnAdtLV9bNwGmSJpcRWKeVsoi+Z/sOYOuw4vlU9/8or+9rKF/uyp3A0H3AiJ7SSlfVO4EPAA9Kur+U/RWwBLhe0kLgKeDssm81VRN8PVUz/HwA21slfQa4u9T79NCN8ogBtct9QElj3Qfc1OH4IkbVyqiq79L8/gTA3Cb1DVwwwmctA5aNN5aIAZH7fdEX8uR4ROfVvQ+4m9zvi27KXFURnTd0H3AJu98H/LCkFVQPyQ7dB4yordlIzYmaYyuJI6KNJF1L9ZDrIZI2Us2SUOs+YESvSeKIaCPb546wq9Z9wIheknscERFRSxJHRETUksQRERG1JHFEREQtuTkeEbvJUqkxmrQ4IiKilrQ4IiJeISaqJZkWR0RE1JLEERERtSRxRERELUkcERFRSxJHRETUksQRERG1JHFEREQtA/kcR556jYhon4FMHHsiySUiYnzSVRUREbX0TItD0jzgi8Ak4ErbS7oc0m7auYbvK0n+O46uH66FeGXricQhaRLwZeBUYCNwt6RVth/pbmRRV5JCa3ItRD/oicQBnACst/0EgKQVwHyg7y6WZv9wDtLvi7YbmGshBlevJI4jgacbtjcCJ3YygPylvKt+TUh7EneP/3/t+rUQMRbZ7nYMSDobON32h8r2B4ATbH9kWL1FwKKy+SbgBx0NdOIdAjzX7SAmWL+f0+ttT+vWL8+1MHD6+bxGvBZ6pcWxETiqYXs68MzwSravAK7oVFDtJmmd7TndjmMiDeI5dViuhQEyqOfVK8Nx7wZmS5olaW/gHGBVl2OK6IZcC9HzeqLFYXuHpA8DN1MNQVxm++EuhxXRcbkWoh/0ROIAsL0aWN3tODpsYLoaGgziOXVUroWBMpDn1RM3xyMion/0yj2OiIjoE0kcERFRSxJHRETUksQRE0rSod2OIaIXDPK1kMTRIZJeJ2mppC9Lmirpv0h6UNL1kg7vdnzjIWnKsJ+pwF2SJkua0u34ojflWuh/GVXVIZJuAm4EDgD+I3ANcC3VBHbvsj2/i+GNi6TfAD8cVjyd6uln235D56OKXpdrof8lcXSIpPtsH1feP2V7RsO++20f273oxkfSJ4B3AX9p+8FS9qTtWd2NLHpZroX+l66qzmn8b718lH19w/bngQ8BfyPpMkkHAvlLJMaSa6HP9eX/pD51g6TXANj+T0OFko4G/qVrUbXI9kbbZwO3AmuA/bscUvS+XAt9Ll1VHSTpjcB/oJr9dAfwOHCt7e1dDawFw87pJeBJYHk/n1O0X66F/pYWR4dI+iiwFNgX+F1gP6ov2PckndzF0MatnNNX2HlO+wKvo4/PKdov10L/S4ujQyQ9CBxr+yVJ+wOrbZ8saQZww9DNwn4yiOcU7TeI35tBPKfRpMXRWUOzEe8DHAhg+yng1V2LqHWDeE7RfoP4vRnEc2qqZ6ZVfwW4Erhb0p3A7wOXAkiaBmztZmAtGMRzivYbxO/NIJ7TiNJV1UGS3gL8NvCQ7ce6Hc9EGMRzivYbxO/NIJ7TSJI4IiKiltzjiIiIWpI4IiKiliSOiIioJYkjIiJqSeKIiIha/j8eWYgPh9wK7QAAAABJRU5ErkJggg==\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "cons.conf.idx\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "euribor3m\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "nr.employed\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "for column in data.select_dtypes(include=['object']).columns:\n", " if column != 'y':\n", " display(pd.crosstab(index=data[column], columns=data['y'], normalize='columns'))\n", "\n", "for column in data.select_dtypes(exclude=['object']).columns:\n", " print(column)\n", " hist = data[[column, 'y']].hist(by='y', bins=30)\n", " plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "다음을 확인할 수 있습니다.\n", "\n", "* \"blue-collar\", \"married\", 디폴트 여부가 \"unknown\"인 사용자가 \"telephone\"으로 \"may\" 시점에 연락되었다면 정기예금 가입에 매우 적은 \"yes\"결과를 얻었을 것입니다.\n", "* 숫자형 변수값의 분포는 정기예금 가입 \"yes\", \"no\"여부에 따라 다르지만 그 관계가 명확하지는 않습니다. \n", "\n", "다음은 속성들이 서로 어떻게 연관되는지 살펴봅니다.\n", "\n" ] }, { "cell_type": "code", "execution_count": 9, "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", "
agedurationcampaignpdayspreviousemp.var.ratecons.price.idxcons.conf.idxeuribor3mnr.employed
age1.000000-0.0008660.004594-0.0343690.024365-0.0003710.0008570.1293720.010767-0.017725
duration-0.0008661.000000-0.071699-0.0475770.020640-0.0279680.005312-0.008173-0.032897-0.044703
campaign0.004594-0.0716991.0000000.052584-0.0791410.1507540.127836-0.0137330.1351330.144095
pdays-0.034369-0.0475770.0525841.000000-0.5875140.2710040.078889-0.0913420.2968990.372605
previous0.0243650.020640-0.079141-0.5875141.000000-0.420489-0.203130-0.050936-0.454494-0.501333
emp.var.rate-0.000371-0.0279680.1507540.271004-0.4204891.0000000.7753340.1960410.9722450.906970
cons.price.idx0.0008570.0053120.1278360.078889-0.2031300.7753341.0000000.0589860.6882300.522034
cons.conf.idx0.129372-0.008173-0.013733-0.091342-0.0509360.1960410.0589861.0000000.2776860.100513
euribor3m0.010767-0.0328970.1351330.296899-0.4544940.9722450.6882300.2776861.0000000.945154
nr.employed-0.017725-0.0447030.1440950.372605-0.5013330.9069700.5220340.1005130.9451541.000000
\n", "
" ], "text/plain": [ " age duration campaign pdays previous \\\n", "age 1.000000 -0.000866 0.004594 -0.034369 0.024365 \n", "duration -0.000866 1.000000 -0.071699 -0.047577 0.020640 \n", "campaign 0.004594 -0.071699 1.000000 0.052584 -0.079141 \n", "pdays -0.034369 -0.047577 0.052584 1.000000 -0.587514 \n", "previous 0.024365 0.020640 -0.079141 -0.587514 1.000000 \n", "emp.var.rate -0.000371 -0.027968 0.150754 0.271004 -0.420489 \n", "cons.price.idx 0.000857 0.005312 0.127836 0.078889 -0.203130 \n", "cons.conf.idx 0.129372 -0.008173 -0.013733 -0.091342 -0.050936 \n", "euribor3m 0.010767 -0.032897 0.135133 0.296899 -0.454494 \n", "nr.employed -0.017725 -0.044703 0.144095 0.372605 -0.501333 \n", "\n", " emp.var.rate cons.price.idx cons.conf.idx euribor3m \\\n", "age -0.000371 0.000857 0.129372 0.010767 \n", "duration -0.027968 0.005312 -0.008173 -0.032897 \n", "campaign 0.150754 0.127836 -0.013733 0.135133 \n", "pdays 0.271004 0.078889 -0.091342 0.296899 \n", "previous -0.420489 -0.203130 -0.050936 -0.454494 \n", "emp.var.rate 1.000000 0.775334 0.196041 0.972245 \n", "cons.price.idx 0.775334 1.000000 0.058986 0.688230 \n", "cons.conf.idx 0.196041 0.058986 1.000000 0.277686 \n", "euribor3m 0.972245 0.688230 0.277686 1.000000 \n", "nr.employed 0.906970 0.522034 0.100513 0.945154 \n", "\n", " nr.employed \n", "age -0.017725 \n", "duration -0.044703 \n", "campaign 0.144095 \n", "pdays 0.372605 \n", "previous -0.501333 \n", "emp.var.rate 0.906970 \n", "cons.price.idx 0.522034 \n", "cons.conf.idx 0.100513 \n", "euribor3m 0.945154 \n", "nr.employed 1.000000 " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "display(data.corr())\n", "pd.plotting.scatter_matrix(data, figsize=(12, 12))\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Notice that:\n", "다음을 확인할 수 있습니다.\n", "\n", "* 속성간 상호 연관성은 다양합니다. 일부는 매우 음의 상관관계를, 또 다른 일부는 양의 상관관계를 보여줍니다. \n", "* 속성간의 관계는 대부분 비선형이며, 연관성이 크지 않습니다." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 변형\n", "\n", "데이터 클린징은 대부분의 머신러닝 프로젝트에서 필요한 작업입니다. 이 작업은 적절히 수행되지 않으면 결과에 악영향을 끼치며, 주관적인 판단이 많이 개입됩니다. 몇가지 일반적인 기술들은 다음과 같습니다.\n", "\n", "* 결측치의 처리 : 일부 머신러닝 알고리즘은 결측치를 처리할 수 있는 경우도 있지만 대부분은 그렇지 않습니다. 이를 처리하는 옵션은:\n", " * 결측값 제거 : 결측값이 매우 일부분일 경우 적용합니다.\n", " * 결측속성 제거 : 다량의 결측값을 가지는 속성이 일부분일 경우 적용합니다.\n", " * 결측값 채우기(imputing) : 다음 책[books](https://www.amazon.com/Flexible-Imputation-Missing-Interdisciplinary-Statistics/dp/1439868247) 전체에서 이 주제에 대해 다루고 있습니다. 일반적인 선택은 결측값을 해당 속성의 다른 값들의 평균이나 최빈값(mode)으로 대체하는 것입니다.\n", "* 명목형(categorical) 속성을 수치형 속성으로 변환 : 가장 일반적인 방법은 원 핫 인코딩(one hot encoding)이라 불리는, 각 명목값들을 컬럼으로 정의한 후 해당값에 매칭되는 여부에 따라 1 또는 0의 값을 가지도록 변환하는 것입니다.\n", "* 분포가 고르지 않은 데이터 : Gradient Boosted Trees와 같은 비선형 모델에서도 좋지 않은 영향을 가져오며, 회기(regression)와 같은 파라미터 방식에서도 과도하게 편향된 데이터는 정확도가 떨어지는 결과를 리턴할 수 있습니다. 간혹 로그(log)값을 취하는 것으로 충분히 정규분포로 변환하는 경우도 있고 개별 범위로 구분하여 명목형 번수로 변환한 후 다시 원 핫 인코딩으로 적용할 수도 있습니다.\n", "* 보다 복잡한 데이터 타입 처리 : 본 노트북에서 다루지는 않지만 이미지, 텍스트, 또는 다양한 grain을 가지는 데이터들에 대해서도 추가 변형이 필요합니다. \n", "\n", "다행히 이들 중 일부는 이미 처리되어 있습니다. 그리고 지금 우리가 다루려고 하는 알고리즘은 드문드문하거나(sparce) 분포가 일정하지 않은 경우에도 잘 동작하는 경향이 있습니다. 따라서 본 예제에서는 최소한의 전처리만 하겠습니다.\n" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "data['no_previous_contact'] = np.where(data['pdays'] == 999, 1, 0) # Indicator variable to capture when pdays takes a value of 999\n", "data['not_working'] = np.where(np.in1d(data['job'], ['student', 'retired', 'unemployed']), 1, 0) # Indicator for individuals not actively employed\n", "model_data = pd.get_dummies(data) # Convert categorical variables to sets of indicators" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "모델을 만들기 전에 필요한 또 다른 질문은 특정 속성이 최종 목표에 기여를 하는지 여부입니다. 예를 들어, 최선의 예측을 제공하는 것이 목표인 경우, 예측을 하려는 시점에 해당 데이터를 사용가능한 지 생각해 봅니다. 우산 판매의 예측에서 비가 올지 여부를 안다면 판매예측에서 매우 유리할 것이지만, 미래의 날씨를 예측하는 것은 날씨 정보 없이 우산판매를 예측하는 것보다 더 어려울 수 있습니다. 이런 경우 과거 날씨정보가 모델의 속성에 포함된다면 정확성을 왜곡할 수도 있습니다. \n", "\n", "이런 논리로, 데이터의 속성들 중 미래에 대한 예측이 필요한 경제 지표들과 `duration`부분을 제외하겠습니다.\n", "\n", "이전 분기의 경제 지표 값을 사용할 수도 있겠지만, 이 값들은 실제 업무환경에서는 현실성이 없을 가능성이 높습니다." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "model_data = model_data.drop(['duration', 'emp.var.rate', 'cons.price.idx', 'cons.conf.idx', 'euribor3m', 'nr.employed'], axis=1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "모델을 만들 때 1차적인 목표는 새로운 데이터에 대한 타겟변수 값을 예측하는 것이며, 이 때 과적합(overfitting)을 이해하는 것이 중요합니다. 지도학습(Supervised learning) 모델은 타겟변수 값에 대한 실제값과 모델의 예측값 사이의 차이(error)를 최소하는 방식으로 설계됩니다. 이 마지막 부분이 중요합니다. 종종 머신러닝 모델은 보다 높은 정확도를 찾는 과정에서 자신이 본 데이터들만에 대한 사소한 특징들까지 고려하는 편중(bias)를 가지게 됩니다. 이런 특징이 새로운 데이터에서 반복적으로 나타나지 않을 경우 실제 예측에서는 정확도가 떨어지게 되고 학습과정에서의 정확도 수준을 보이지 않게 됩니다. \n", "\n", "이를 예방하는 가장 일반적인 방법은 모델이 학습을 할 때 학습데이터 뿐 아니라 새로운 데이터에 대해서도 적합성을 함께 판단하도록 하는 것이며 홀드아웃 검증(holdout validation), 교차검증(cross-validation), 일회성 검증(leave-one-out validation) 등 여러가지 방식이 있습니다. 본 예제에서는 단순히 랜덤하게 3개의 그룹으로 데이터를 나눌 것입니다. 모델은 70%의 데이터를 이용하여 학습을 하고, 20%의 데이터를 새로운 데이터에대한 정확도를 평가하는 용도로 사용하고, 10%의 데이터를 마지막 테스트셋으로 분리하여 성능을 테스트하겠습니다. 또한 데이터셋 분리시 랜덤하게 순서를 조정하고 있음에도 주목합니다.\n" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "train_data, validation_data, test_data = np.split(model_data.sample(frac=1, random_state=1729), [int(0.7 * len(model_data)), int(0.9 * len(model_data))]) # Randomly sort the data then split out first 70%, second 20%, and last 10%" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Amazon SageMaker의 XGBoost 컨테이너는 libSVM 또는 CSV 포맷의 데이터를 사용합니다. 본 예제에서는 CSV를 이용합니다. CSV파일에서 첫번째 컬럼을 타겟변수 값으로 지정해야 하며 헤더를 포함하고 있지 않아야 합니다. 본 예제에서는 데이터를 train|validation|test 데이터셋으로 분리한 후 작업을 하고 있습니다. " ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "pd.concat([train_data['y_yes'], train_data.drop(['y_no', 'y_yes'], axis=1)], axis=1).to_csv('train.csv', index=False, header=False)\n", "pd.concat([validation_data['y_yes'], validation_data.drop(['y_no', 'y_yes'], axis=1)], axis=1).to_csv('validation.csv', index=False, header=False)\n", "pd.concat([test_data['y_yes'], test_data.drop(['y_no', 'y_yes'], axis=1)], axis=1).to_csv('test.csv', index=False, header=False)\n", "pd.concat([test_data.drop(['y_no', 'y_yes'], axis=1)], axis=1).to_csv('test_features.csv', index=False, header=False)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "다음은 SageMaker의 관리형 학습환경에서 이 데이터에 접근할 수 있도록 파일을 S3로 복사하겠습니다." ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "boto3.Session().resource('s3').Bucket(bucket).Object(os.path.join(prefix, 'train/train.csv')).upload_file('train.csv')\n", "boto3.Session().resource('s3').Bucket(bucket).Object(os.path.join(prefix, 'validation/validation.csv')).upload_file('validation.csv')\n", "boto3.Session().resource('s3').Bucket(bucket).Object(os.path.join(prefix, 'test/test.csv')).upload_file('test.csv')\n", "boto3.Session().resource('s3').Bucket(bucket).Object(os.path.join(prefix, 'test/test_features.csv')).upload_file('test.csv')\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "\n", "## 학습\n", "\n", "우리가 사용하는 데이터의 많은 속성들이 편향된 분포를 가지고 있습니다. 일부 속성들은 서로 높은 연관성을 가지고 있고 일부는 타겟속성 값과 비선형 관계를 가지고 있었습니다. 또한 미래 마케팅에 대한 예측에서 높은 정확도가 필요하고 왜 그렇게 판단하는지에 대한 설명 또한 중요합니다. 이런 점들을 고려할 때 Gradient boosted tree와 같은 알고리즘이 매우 적합한 후보입니다. \n", "\n", "Gradient boosted tree는 작은 모델들이 결합되어 작동하며, 각 모델은 이전 모델의 결함을 보완하는 방식으로 동작합니다. 단순한 모델들이 모여 크고 복잡한 다른 모델들보다 높은 성능을 냅니다. Gradient boosting tree알고리즘이 다른 알고리즘과 어떻게 다른지에 대해 설명하는 다른 SageMaker 노트북이 있으니 이를 참고합니다.\n", "\n", "`xgboost`는 매우 인기있는 Gradient bossted tree에 대한 오픈소스 패키지 입니다. 계산성능이 뛰어나고, 필요한 기능들을 모두 구현하고 있으며, 많은 머신러닝 경쟁에서 성공적인 성과를 보여주고 있습니다. SageMaker의 관리형, 분산 학습 프레임워크를 이용하여 학습할 수 있도록 간단한 `xgboost`모델을 시작해 보겠습니다. \n", "\n", "먼저, SageMaker의 XGBoost구현체가 있는 ECR 컨터이너를 지정합니다." ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "from sagemaker import image_uris \n", "container = image_uris.retrieve('xgboost', region='us-east-1', version='latest')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "우리는 CSV 파일 포맷을 사용하므로 S3의 파일 위치를 알려주는 `s3_input`오브젝트를 생성하고 콘텐츠 타입을 CSV로 지정합니다." ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [], "source": [ "s3_input_train = sagemaker.inputs.TrainingInput(s3_data='s3://{}/{}/train'.format(bucket, prefix), content_type='csv')\n", "s3_input_validation = sagemaker.inputs.TrainingInput(s3_data='s3://{}/{}/validation/'.format(bucket, prefix), content_type='csv')\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "다음으로 다음 파라미터를 지정하여 esitmator를 생성합니다.\n", "\n", "1. `xgboost` 알고리즘 컨테이너를 사용\n", "1. 사용할 IAM 역할(role)\n", "1. 학습용 인스턴스 타입과 수량 \n", "1. 출력데이터를 위한 S3위치 \n", "1. 알고리즘 하이퍼파라미터 \n", "\n", "이제 다음 파라미터를 이용하여 `.fit()` 명령을 실행합니다.\n", "1. 학습용 데이터가 있는 S3 위치. 본 예제는 학습과 검증 데이터셋을 모두 사용하므로 두 채널을 모두 지정합니다.\n" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "2021-01-23 06:34:42 Starting - Starting the training job...\n", "2021-01-23 06:35:07 Starting - Launching requested ML instancesProfilerReport-1611383682: InProgress\n", "......\n", "2021-01-23 06:36:08 Starting - Preparing the instances for training......\n", "2021-01-23 06:37:11 Downloading - Downloading input data\n", "2021-01-23 06:37:11 Training - Downloading the training image...\n", "2021-01-23 06:37:36 Uploading - Uploading generated training model\u001b[34mArguments: train\u001b[0m\n", "\u001b[34m[2021-01-23:06:37:31:INFO] Running standalone xgboost training.\u001b[0m\n", "\u001b[34m[2021-01-23:06:37:31:INFO] File size need to be processed in the node: 4.35mb. Available memory size in the node: 8417.98mb\u001b[0m\n", "\u001b[34m[2021-01-23:06:37:31:INFO] Determined delimiter of CSV input is ','\u001b[0m\n", "\u001b[34m[06:37:31] S3DistributionType set as FullyReplicated\u001b[0m\n", "\u001b[34m[06:37:31] 28831x59 matrix with 1701029 entries loaded from /opt/ml/input/data/train?format=csv&label_column=0&delimiter=,\u001b[0m\n", "\u001b[34m[2021-01-23:06:37:31:INFO] Determined delimiter of CSV input is ','\u001b[0m\n", "\u001b[34m[06:37:31] S3DistributionType set as FullyReplicated\u001b[0m\n", "\u001b[34m[06:37:31] 8238x59 matrix with 486042 entries loaded from /opt/ml/input/data/validation?format=csv&label_column=0&delimiter=,\u001b[0m\n", "\u001b[34m[06:37:31] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[0]#011train-error:0.100482#011validation-error:0.103545\u001b[0m\n", "\u001b[34m[06:37:31] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 16 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[1]#011train-error:0.099858#011validation-error:0.103545\u001b[0m\n", "\u001b[34m[06:37:31] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 18 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[2]#011train-error:0.099476#011validation-error:0.10403\u001b[0m\n", "\u001b[34m[06:37:31] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[3]#011train-error:0.099025#011validation-error:0.10403\u001b[0m\n", "\u001b[34m[06:37:31] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 18 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[4]#011train-error:0.099476#011validation-error:0.10318\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 8 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[5]#011train-error:0.099372#011validation-error:0.10318\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 12 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[6]#011train-error:0.09906#011validation-error:0.10318\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 20 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[7]#011train-error:0.099025#011validation-error:0.102938\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 24 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[8]#011train-error:0.099164#011validation-error:0.102816\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 12 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[9]#011train-error:0.098817#011validation-error:0.103666\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 30 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[10]#011train-error:0.098817#011validation-error:0.103787\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 22 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[11]#011train-error:0.098817#011validation-error:0.103545\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 22 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[12]#011train-error:0.098852#011validation-error:0.103545\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 16 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[13]#011train-error:0.098574#011validation-error:0.103666\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 18 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[14]#011train-error:0.098609#011validation-error:0.10403\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 8 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[15]#011train-error:0.098401#011validation-error:0.103909\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 18 extra nodes, 26 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[16]#011train-error:0.098401#011validation-error:0.10403\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 22 extra nodes, 20 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[17]#011train-error:0.098297#011validation-error:0.103545\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 18 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[18]#011train-error:0.098054#011validation-error:0.103545\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 20 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[19]#011train-error:0.098158#011validation-error:0.10318\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 22 extra nodes, 8 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[20]#011train-error:0.098193#011validation-error:0.103787\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 36 extra nodes, 8 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[21]#011train-error:0.098193#011validation-error:0.103302\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 28 extra nodes, 22 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[22]#011train-error:0.098124#011validation-error:0.10318\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 16 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[23]#011train-error:0.098124#011validation-error:0.103545\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 20 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[24]#011train-error:0.097881#011validation-error:0.103909\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 34 extra nodes, 4 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[25]#011train-error:0.097777#011validation-error:0.104273\u001b[0m\n", "\u001b[34m[06:37:32] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 22 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[26]#011train-error:0.097742#011validation-error:0.104151\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 20 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[27]#011train-error:0.097707#011validation-error:0.104394\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 22 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[28]#011train-error:0.097291#011validation-error:0.104394\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 32 extra nodes, 4 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[29]#011train-error:0.097152#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 6 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[30]#011train-error:0.097256#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 24 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[31]#011train-error:0.097083#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 16 extra nodes, 10 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[32]#011train-error:0.097083#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 16 extra nodes, 24 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[33]#011train-error:0.097083#011validation-error:0.10488\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 18 extra nodes, 8 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[34]#011train-error:0.097152#011validation-error:0.10488\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 20 extra nodes, 28 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[35]#011train-error:0.097256#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 22 extra nodes, 22 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[36]#011train-error:0.097187#011validation-error:0.104394\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 26 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[37]#011train-error:0.097118#011validation-error:0.104516\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 18 extra nodes, 20 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[38]#011train-error:0.097152#011validation-error:0.104516\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 20 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[39]#011train-error:0.096736#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 4 extra nodes, 14 pruned nodes, max_depth=2\u001b[0m\n", "\u001b[34m[40]#011train-error:0.09691#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 18 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[41]#011train-error:0.096736#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 20 extra nodes, 16 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[42]#011train-error:0.096771#011validation-error:0.10488\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 12 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[43]#011train-error:0.096806#011validation-error:0.10488\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 18 extra nodes, 16 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[44]#011train-error:0.096736#011validation-error:0.105001\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 4 extra nodes, 14 pruned nodes, max_depth=2\u001b[0m\n", "\u001b[34m[45]#011train-error:0.096806#011validation-error:0.105123\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 22 extra nodes, 24 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[46]#011train-error:0.096459#011validation-error:0.104516\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 26 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[47]#011train-error:0.096424#011validation-error:0.104394\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 16 pruned nodes, max_depth=4\u001b[0m\n", "\u001b[34m[48]#011train-error:0.096528#011validation-error:0.104273\u001b[0m\n", "\u001b[34m[06:37:33] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 20 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[49]#011train-error:0.096563#011validation-error:0.103666\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 20 extra nodes, 38 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[50]#011train-error:0.096597#011validation-error:0.10403\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[51]#011train-error:0.096528#011validation-error:0.104394\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 16 extra nodes, 30 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[52]#011train-error:0.096112#011validation-error:0.104394\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 16 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[53]#011train-error:0.096077#011validation-error:0.104394\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 16 extra nodes, 16 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[54]#011train-error:0.09632#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 8 extra nodes, 16 pruned nodes, max_depth=3\u001b[0m\n", "\u001b[34m[55]#011train-error:0.09632#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 30 pruned nodes, max_depth=4\u001b[0m\n", "\u001b[34m[56]#011train-error:0.096147#011validation-error:0.104516\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 16 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[57]#011train-error:0.09632#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 22 extra nodes, 24 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[58]#011train-error:0.096112#011validation-error:0.104394\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 34 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[59]#011train-error:0.096042#011validation-error:0.104273\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 28 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[60]#011train-error:0.096008#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 26 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[61]#011train-error:0.096042#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 20 extra nodes, 30 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[62]#011train-error:0.096077#011validation-error:0.104516\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 6 extra nodes, 30 pruned nodes, max_depth=3\u001b[0m\n", "\u001b[34m[63]#011train-error:0.096147#011validation-error:0.104273\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 18 extra nodes, 18 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[64]#011train-error:0.096216#011validation-error:0.10403\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 20 extra nodes, 12 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[65]#011train-error:0.09632#011validation-error:0.104151\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 10 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[66]#011train-error:0.096181#011validation-error:0.104273\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 20 pruned nodes, max_depth=4\u001b[0m\n", "\u001b[34m[67]#011train-error:0.095904#011validation-error:0.104151\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 16 extra nodes, 24 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[68]#011train-error:0.096008#011validation-error:0.104516\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 26 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[69]#011train-error:0.096042#011validation-error:0.104516\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 22 extra nodes, 12 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[70]#011train-error:0.096077#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 8 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[71]#011train-error:0.095938#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 30 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[72]#011train-error:0.095938#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 4 extra nodes, 28 pruned nodes, max_depth=2\u001b[0m\n", "\u001b[34m[73]#011train-error:0.096008#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 12 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[74]#011train-error:0.095869#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 34 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[75]#011train-error:0.095938#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 24 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[76]#011train-error:0.095904#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:34] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[77]#011train-error:0.0958#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 6 extra nodes, 14 pruned nodes, max_depth=3\u001b[0m\n", "\u001b[34m[78]#011train-error:0.09573#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 18 pruned nodes, max_depth=4\u001b[0m\n", "\u001b[34m[79]#011train-error:0.095765#011validation-error:0.105123\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 14 extra nodes, 12 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[80]#011train-error:0.095834#011validation-error:0.104637\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 20 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[81]#011train-error:0.095592#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 18 extra nodes, 22 pruned nodes, max_depth=4\u001b[0m\n", "\u001b[34m[82]#011train-error:0.095557#011validation-error:0.104394\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 6 extra nodes, 26 pruned nodes, max_depth=3\u001b[0m\n", "\u001b[34m[83]#011train-error:0.095557#011validation-error:0.104273\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 32 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[84]#011train-error:0.095453#011validation-error:0.104758\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 8 pruned nodes, max_depth=4\u001b[0m\n", "\u001b[34m[85]#011train-error:0.095453#011validation-error:0.105001\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 24 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[86]#011train-error:0.095453#011validation-error:0.10488\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 24 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[87]#011train-error:0.095453#011validation-error:0.10488\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 20 extra nodes, 16 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[88]#011train-error:0.095349#011validation-error:0.10488\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 4 extra nodes, 30 pruned nodes, max_depth=2\u001b[0m\n", "\u001b[34m[89]#011train-error:0.095037#011validation-error:0.105365\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 14 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[90]#011train-error:0.095106#011validation-error:0.105487\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 42 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[91]#011train-error:0.095037#011validation-error:0.105487\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 0 extra nodes, 30 pruned nodes, max_depth=0\u001b[0m\n", "\u001b[34m[92]#011train-error:0.095106#011validation-error:0.105365\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 10 extra nodes, 14 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[93]#011train-error:0.095314#011validation-error:0.10488\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 24 extra nodes, 30 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[94]#011train-error:0.095314#011validation-error:0.105123\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 6 extra nodes, 24 pruned nodes, max_depth=3\u001b[0m\n", "\u001b[34m[95]#011train-error:0.095314#011validation-error:0.105123\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 12 extra nodes, 30 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[96]#011train-error:0.095279#011validation-error:0.105123\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 18 extra nodes, 12 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[97]#011train-error:0.094828#011validation-error:0.105487\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 4 extra nodes, 22 pruned nodes, max_depth=2\u001b[0m\n", "\u001b[34m[98]#011train-error:0.094863#011validation-error:0.105365\u001b[0m\n", "\u001b[34m[06:37:35] src/tree/updater_prune.cc:74: tree pruning end, 1 roots, 30 extra nodes, 12 pruned nodes, max_depth=5\u001b[0m\n", "\u001b[34m[99]#011train-error:0.094759#011validation-error:0.104758\u001b[0m\n", "\n", "2021-01-23 06:38:09 Completed - Training job completed\n", "Training seconds: 49\n", "Billable seconds: 49\n" ] } ], "source": [ "sess = sagemaker.Session()\n", "\n", "xgb = sagemaker.estimator.Estimator(container,\n", " role, \n", " instance_count=1, \n", " instance_type='ml.m4.xlarge',\n", " output_path='s3://{}/{}/output'.format(bucket, prefix),\n", " sagemaker_session=sess)\n", "xgb.set_hyperparameters(max_depth=5,\n", " eta=0.2,\n", " gamma=4,\n", " min_child_weight=6,\n", " subsample=0.8,\n", " silent=0,\n", " objective='binary:logistic',\n", " num_round=100)\n", "\n", "xgb.fit({'train': s3_input_train, 'validation': s3_input_validation}) " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "\n", "## 호스팅\n", "\n", "입력데이터에 대해 `xgboost` 모델의 학습이 완료되면 이 모델을 실시간 추론을 위한 엔드포인트로 배포하겠습니다." ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "---------------!" ] } ], "source": [ "xgb_predictor = xgb.deploy(initial_instance_count=1,\n", " instance_type='ml.m4.xlarge')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "\n", "## 평가\n", "\n", "머신러닝 모델의 성능을 확인하는 여러가지 방법이 있습니다. 여기서는 단순히 실제값과 예측값을 비교하겠습니다. 정기예금에 가입을 한 경우(`1`) 와 그렇지 않은 경우(`0`)를 이용하여 혼돈행렬(confusion matrix)를 생성하겠습니다.\n", "\n", "이를 위해 추론용 데이터를 엔드포인트에 전달하고 결과를 받아야 합니다. 현재 데이터는 노트북 인스턴스의 메모리에 NumPy 배열로 저장되어 있습니다. 데이터를 HTTP POST request로 보내기 위해 CSV형태로 직렬화(serialize)하고 결과로 리턴되는 CSV를 디코딩합니다.\n", "\n", "*주의: SageMaker XGBoost에서 CSV포맷으로 추론할 때 요청 데이터는 타겟속성 컬럼을 포함하지 않습니다.*" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [], "source": [ "from sagemaker.serializers import CSVSerializer\n", "xgb_predictor.serializer = CSVSerializer()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "엔드포인트를 호출하는 간단한 함수를 생성합니다.:\n", "\n", "1. 테스트 데이터셋을 반복(Loop)\n", "1. rows 만큼 미니매치로 나누기\n", "1. 미니배치를 CSV string payloads로 변환 (타겟속성 변수를 제거합니다.)\n", "1. XGBoost 엔드포인트를 호출하고 예측값 수신\n", "1. CSV결과로 리턴된 예측값을 다시 NumPy 배열로 변환\n" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [], "source": [ "def predict(data, rows=500):\n", " split_array = np.array_split(data, int(data.shape[0] / float(rows) + 1))\n", " predictions = ''\n", " for array in split_array:\n", " predictions = ','.join([predictions, xgb_predictor.predict(array).decode('utf-8')])\n", "\n", " return np.fromstring(predictions[1:], sep=',')\n", "\n", "predictions = predict(test_data.drop(['y_no', 'y_yes'], axis=1).to_numpy())" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "예측결과와 실제값을 비교하는 혼돈행렬을 생성합니다." ] }, { "cell_type": "code", "execution_count": 21, "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", "
predictions0.01.0
actuals
0359442
138994
\n", "
" ], "text/plain": [ "predictions 0.0 1.0\n", "actuals \n", "0 3594 42\n", "1 389 94" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "pd.crosstab(index=test_data['y_yes'], columns=np.round(predictions), rownames=['actuals'], colnames=['predictions'])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "모델의 예측결과, 약 4,000명의 잠재고객에 대하여 136명이 정기예금에 가입할 것으로 예측하였고 이 중 94명이 실제로 가입한 것으로 확인됩니다. 그리고 모델이 가입할 것으로 예측하지 않았으나 실제로 가입한 고객은 389명으로 확인됩니다. 이 결과는 기대했던 것보다 낮은 성능일 수 있습니다. 하지만 우리는 최소의 노력으로 링크[here](http://media.salford-systems.com/video/tutorial/2015/targeted_marketing.pdf)에 소개된 결과와 유사한 수준의 정확도를 달성하였고 또 더 개선될 여지도 있습니다. \n", "\n", "_알고리즘의 샘플링과정에서 랜덤요소가 반영되므로 결과의 숫자는 위 결과와 정확히 동일하지 않을 수 있습니다._" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "\n", "## 확장\n", "\n", "본 예제는 비교적 작은 데이터셋을 이용한 분석이지만 SageMaer의 분산, 관리형 학습과 실시간 모델 호스팅 기능은 대량의 데이터를 다루어야 하는 문제에도 쉽게 적용될 수 있습니다. 예측 정확도를 더 개선하기 위해 false-positives와 false-negatives에 변화를 주도록 threshold값을 조정할 수 있습니다. 실제 업무환경에서는 데이터의 속성을 보다 면밀히 살피고, 현재 데이터셋에서 추가로 더 많은 고객정보를 확보하기 위해 더 많은 시간을 소비하게 될 것입니다." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### (옵션) 리소스 제거\n", "\n", "본 예제를 모두 마무리한 후 아래 셀을 실행합니다. 다음 명령은 추론 단계에서 생성한 SageMaker에서 호스팅되고 있는 엔드포인트를 제거합니다. 엔드포인트를 삭제하지 않으면 계속 사용요금이 발생할 수 있습니다.\n" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "xgb_predictor.delete_endpoint()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "celltoolbar": "Tags", "kernelspec": { "display_name": "conda_python3", "language": "python", "name": "conda_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": "3.6.10" }, "notice": "Copyright 2017 Amazon.com, Inc. or its affiliates. All Rights Reserved. Licensed under the Apache License, Version 2.0 (the \"License\"). You may not use this file except in compliance with the License. A copy of the License is located at http://aws.amazon.com/apache2.0/ or in the \"license\" file accompanying this file. This file is distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License." }, "nbformat": 4, "nbformat_minor": 4 }