Skip to content

Commit

Permalink
merging
Browse files Browse the repository at this point in the history
  • Loading branch information
ucrystal committed Apr 16, 2019
1 parent 442b3fe commit ea85dd6
Show file tree
Hide file tree
Showing 213 changed files with 11,511 additions and 3,557 deletions.
101 changes: 101 additions & 0 deletions docs/_downloads/sales.csv

Large diffs are not rendered by default.

Binary file added docs/_images/create_a_dataset_of_database_4.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_a_dataset_of_stagingdb_2.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_db_1.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_db_10.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_db_11.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_db_12.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_db_3.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_db_5.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_db_5_2.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_db_9.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_file_2.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_file_5.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_file_6.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_file_7.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_file_8.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_file_9.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_stagingdb_1.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_stagingdb_5.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/create_datasource_stagingdb_6.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/discovery-main.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/s1-1.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/s1-10.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/s1-11.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/s1-2.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/_images/s1-3.png
Binary file added docs/_images/s1-4.png
Binary file added docs/_images/s1-5.png
Binary file added docs/_images/s1-6.png
Binary file added docs/_images/s1-7.png
Binary file added docs/_images/s1-8.png
Binary file added docs/_images/s1-9.png
Binary file added docs/_images/tt01.png
Binary file added docs/_images/tt02.png
Binary file added docs/_images/tt03-2.png
Binary file added docs/_images/tt03-3.png
Binary file added docs/_images/tt03.png
Binary file added docs/_images/tt04.png
Binary file added docs/_images/tt05.png
Binary file added docs/_images/tt06.png
Binary file added docs/_images/tt07.gif
Binary file added docs/_images/tt08.gif
25 changes: 22 additions & 3 deletions docs/_sources/discovery/part00/index.rst.txt
@@ -1,10 +1,29 @@
디스커버리 시작하기
---------------------------

Metatron Discovery는 대용량 데이터를 동일한 환경에서 적재, 전처리, 분석까지 빠르게 처리할 수 있는 툴입니다. 데이터 분석을 위한 기술적인 지식이 없는 비즈니스 현업 사용자도 Metatron Discovery를 이용하면 직접 데이터를 다뤄보고 빠르게 시각화해 보면서 인사이트를 얻을 수 있습니다.

Metatron Discovery로 즉시 데이터 분석을 시작할 수 있는 방법은 두 가지가 있습니다.

1. `Metatron Discovery 데모 사이트 <https://metatron.app>`_ 를 실행하세요. 데모를 위한 **ID** 와 **password** 는 ``metatron`` 으로 제공됩니다.
2. 싱글 모드용 Metatron Discovery를 로컬 PC에 다운로드 받으세요. `다운로드 <https://metatron.app/download/>`_ 는 세 가지 방법으로 제공됩니다.

- `Custom install <https://metatron.app/download/installation-guide-custom-install/>`_ : Github 레파지토리로부터 소스코드를 다운받거나 빌드 파일을 직접 실행하세요.
- `Virtual machine <https://metatron.app/download/guide_vm/>`_ : 가상 머신 이미지로 바로 실행하세요. Windows OS에서도 사용할 수 있습니다.
- `Docker <https://metatron.app/download/installation-guide-docker/>`_ : Docker 이미지로 빠르게 설치하고 즉시 실행할 수 있습니다.


.. figure:: /_static/img/discovery/part00/discovery-main.png
:align: center
:alt: Metatron Discovery main page

위와 같은 화면을 보고 계신가요? 축하합니다! 이제부터 메타트론 디스커버리와 함께 쉽고 빠르게 데이터 분석을 시작해보세요.

빠른 시작을 위해서 아래의 세 단계의 스텝으로 튜토리얼을 진행해보세요.

.. toctree::
:maxdepth: 4

../part00/overview

../part00/step1
../part00/step2
../part00/step3
../part00/step3
94 changes: 93 additions & 1 deletion docs/_sources/discovery/part00/step1.rst.txt
@@ -1,2 +1,94 @@
Step1.데이터 소스 만들기
Step1. 데이터 소스 만들기
---------------------------------------------

데이터 분석을 하기 위해 가장 먼저 해야할 일은 데이터를 시스템에 적재하는 것 입니다. Metatron Discovery는 다양한 원천 데이터를 쉽게 적재할 수 있도록 지원합니다.

튜토리얼에서는 내 로컬 파일에 있는 데이터를 적재하는 방법을 소개합니다. 먼저 데이터를 준비하세요. 흔히 쓰이는 엑셀 파일(.xls, .xlsx)또는 .csv 형식의 파일이면 충분합니다. 본 튜토리얼에서는 판매 현황 데이터를 활용합니다. 아래 링크에서 다운로드 받으세요.

:download:`sample data (.csv) </_static/data/sales.csv>`


데이터 소스는 ``Management > 데이터 스토리지 > 데이터 소스`` 에서 조회하고 적재할 수 있습니다. 새로운 데이터 소스를 만들기 위해서 데이터 소스 리스트 우측 상단의 ``new`` 버튼을 클릭합니다.

.. figure:: /_static/img/discovery/part00/s1-1.png
:align: center
:alt: create a new data source

튜토리얼에서는 ``My File`` 을 눌러 내 로컬 폴더에서 데이터를 가져옵니다. 다른 원천에서 데이터 소스를 만드는 방법은 :doc:`../part02/create_a_data_source` 문서를 참조하세요.

분석하고 싶은 데이터를 drag & drop하거나 디렉토리로부터 불러오기 할 수 있습니다.

.. figure:: /_static/img/discovery/part00/s1-2.png
:align: center
:alt: create a data source from local

판매 현황 데이터를 드래그하면 컬럼과 라인 구분자에 따라 나타나는 데이터 샘플을 최대 100행 까지 조회할 수 있습니다. 이 데이터는 기본적으로 설정된 구분자로 충분히 데이터를 잘 나타내는 것 같네요! ``다음`` 을 누릅니다.

.. figure:: /_static/img/discovery/part00/s1-3.png
:align: center
:alt: data preview

이제 실제로 데이터를 보면서 컬럼의 타입을 알맞게 조정해야 합니다. 이 작업을 ``데이터 스키마 구성`` 이라고 부릅니다.

.. figure:: /_static/img/discovery/part00/s1-4.png
:align: center
:alt: data schema configuration

각 컬럼은 차원값(dimension) 또는 측정값(measure) 라는 역할로 나뉩니다. 자세한 내용은 :doc:`../part02/dimension_and_measures` 문서를 참조하세요. 이 데이터에서는 ``Discount`` ``Profit`` ``Quantity`` ``Sales`` ``DaystoShipActual`` ``SalesForecast`` ``DaystoShipScheduled`` ``SalesperCustomer`` ``ProfitRatio`` 컬럼들은 측정값으로 변경해야 합니다.

그 다음은 컬럼의 데이터 타입을 적절하게 변경해 주어야 합니다. 기본적으로 차원값은 문자로, 측정값은 정수로 설정되어 있습니다. 데이터 샘플을 보면서 가장 알맞는 형식으로 변경해주세요. 아래에 이 데이터에서 변경할 사항들을 나열했습니다.

- ``Orderdate`` : 날짜/시간
- ``Discount`` : 소수
- ``ShipDate`` : 날짜/시간 (시간 표현을 yyyy. MM. dd. 로 변경한 후 체크박스 클릭하여 유효성 확인)
- ``SalesperCustomer`` : 소수
- ``ProfitRatio`` : 소수
- ``latitude`` : 위도
- ``longitude`` : 경도

마지막으로 새로운 컬럼을 만들어 줄 차례입니다. 우리는 위도와 경도 컬럼을 갖고 있으므로 Point 타입의 컬럼을 새로 만들 수 있습니다. 우측 상단의 ``컬럼 추가`` 버튼을 누르세요. 위도 컬럼에 ``latitude`` 컬럼을 선택하고, 경도 컬럼에 ``longitude`` 컬럼을 선택합니다. 컬럼 이름을 적절하게 입력한 후 추가를 누르세요. 새로운 ``Point`` 타입 컬럼이 생성되었습니다!

.. figure:: /_static/img/discovery/part00/s1-5.png
:align: center
:alt: create a new Point type column

스키마 구성 작업이 마무리되었으면 ``다음`` 을 누릅니다. 필요한 경우 드루이드의 수집 설정을 변경하는 작업입니다. 지금은 기본 설정으로도 충분합니다.

.. figure:: /_static/img/discovery/part00/s1-6.png
:align: center
:alt: druid ingestion setting

마지막으로 데이터 소스의 이름과 설명을 입력합니다. ``마침`` 을 누르면 즉시 데이터 소스 상세로 넘어갑니다.

.. figure:: /_static/img/discovery/part00/s1-7.png
:align: center
:alt: typing name and description for a new data source in metatron discovery

데이터 소스 상세에서는 적재 현황을 실시간으로 볼 수 있습니다. 몇 분 기다리면 아래와 같이 적재가 성공했음을 알리며 히스토그램이 나타납니다. 혹시 다른 데이터 소스를 적재하다 에러가 났을 경우 ``상세`` 링크를 클릭하여 드루이드 적재 로그를 조회할 수 있습니다. 컬럼 명 중복, 컬럼 타입과 불일치하는 데이터 등으로 인해 적재가 실패할 수 있습니다. 이 경우 원인을 찾아 다시 적재를 시도해보아야 합니다.

.. figure:: /_static/img/discovery/part00/s1-8.png
:align: center
:alt: success for creation of data source for the metatron discovery

만약 이 데이터 소스를 다른 사용자들에게 오픈하려면 ``할당`` 에서 ``모든 워크스페이스에 데이터소스를 사용하도록 허용`` 체크박스에 체크합니다. 전체 유저가 아니라 특정 유저들에게만 오픈하고 싶으면 ``수정`` 을 클릭하여 할당하고자 하는 개인 사용자들 또는 팀 워크스페이스를 선택합니다.

.. figure:: /_static/img/discovery/part00/s1-9.png
:align: center
:alt: assign data source to specific workspaces

이 예제에서는 모든 사용자가 사용할 수 있도록 ``Open Data`` 로 설정하겠습니다.

.. figure:: /_static/img/discovery/part00/s1-10.png
:align: center
:alt: make public data

적재된 데이터는 ``데이터 탭`` 에서 조회할 수 있습니다.

.. figure:: /_static/img/discovery/part00/s1-11.png
:align: center
:alt: data tab in metatron discovery


축하합니다! 이제 데이터 소스를 사용해 볼 차례네요. 다음 스텝으로 넘어가볼까요?

:doc:`../part00/step2`
42 changes: 41 additions & 1 deletion docs/_sources/discovery/part00/step2.rst.txt
@@ -1,2 +1,42 @@
Step2.워크북 만들기
Step2. 워크북 만들기
---------------------------------------------

분석을 위한 데이터가 준비되었나요? 그럼 이제 워크북을 만들 차례입니다. 워크북은 데이터 시각화 기능을 포함하는 컴포넌트입니다. 좌측 상단의 Metatron Discovery 로고를 클릭하면 메인 화면인 내 개인 워크스페이스로 이동합니다.

.. figure:: /_static/img/discovery/part00/tt02.png
:align: center
:alt: Sales data

우측 하단의 ``+ 워크북`` 버튼을 눌러 워크북을 만들어 볼까요? 만들 워크북의 이름과 설명을 입력합니다. 워크북을 만드는 즉시 이어서 대시보드를 만들도록 체크박스에 표시가 되어 있습니다. 여기서 각 워크북은 여러 개의 대시보드를 포함하고, 각 대시보드는 또 여러 개의 차트를 포함하는 구조입니다.

.. figure:: /_static/img/discovery/part00/tt03.png
:align: center
:alt: Sales data

이어서 대시보드를 만들어야 합니다. 대시보드에는 시각화할 데이터 소스가 있어야 합니다. 이 데이터 소스는 단일 소스이거나, join으로 연결된 데이터 소스일 수도 있습니다. 더 자세한 내용은 :doc:`../part04/create_a_dashboard` 문서를 참고하세요. 본 튜토리얼에서는 Step1에서 적재한 ``판매현황 데이터`` 하나만 사용합니다.

.. figure:: /_static/img/discovery/part00/tt03-2.png
:align: center
:alt: Sales data

데이터 소스 추가 버튼을 누르면 데이터 소스 선택 팝업이 등장합니다. ``판매현황 데이터`` 를 검색하거나 오픈 데이터만 보기에 체크하여 공개된 데이터 중에서 찾아냅니다.

.. figure:: /_static/img/discovery/part00/tt04.png
:align: center
:alt: Sales data

마지막으로 대시보드의 이름과 설명을 입력합니다.

.. figure:: /_static/img/discovery/part00/tt03-3.png
:align: center
:alt: Sales data

워크북 내에 대시보드가 생성되었습니다! 이제 이 위젯들을 추가해서 대시보드를 구성하는 일만 남았네요.

.. figure:: /_static/img/discovery/part00/tt06.png
:align: center
:alt: Sales data

다음 단계로 넘어가 볼까요?

:doc:`../part00/step3`
14 changes: 12 additions & 2 deletions docs/_sources/discovery/part00/step3.rst.txt
@@ -1,6 +1,16 @@
Step3.대시보드/차트 만들기
Step3. 대시보드 구성하기
---------------------------------------------

빈 대시보드에 차트 위젯, 텍스트 위젯, 필터 위젯을 넣어 보면서 대시보드를 구성할 차례입니다.

.. figure:: /_static/img/discovery/part00/create_chart.png
:align: center
:alt: 차트 생성
:alt: 차트 생성

.. figure:: /_static/img/discovery/part00/tt07.gif
:align: center
:alt: Sales data

.. figure:: /_static/img/discovery/part00/tt08.gif
:align: center
:alt: Sales data
20 changes: 9 additions & 11 deletions docs/_sources/discovery/part01/metatron_engine.rst.txt
Expand Up @@ -76,12 +76,12 @@ Druid는 컬럼들을 각각 따로 저장한다. Druid가 주로 이벤트 스

Example:
::
Justin Bieber -> 0
Justin Bieber -> 0
Ke$ha -> 1

이 매핑을 사용하면 page 컬럼을 정수 배열로 나타낼 수 있는데, 여기서 배열 인덱스 각각은 원본 데이터셋의 각 행에 해당한다. page 컬럼의 경우, 각 행의 page 값을 아래와 같이 표시할 수 있다.
::
[0, 0, 1, 1]
[0, 0, 1, 1]

이처럼 문자열들이 고정 길이 정수들로 바뀌어 저장되므로 압축하기가 훨씬 더 수월하다. Druid는 각 shard(세그먼트) 단위로 데이터를 인덱싱한다.

Expand All @@ -90,27 +90,25 @@ Example:

Druid는 검색 인덱스를 추가로 만들어서 문자열 컬럼에 대한 필터링을 용이하게 할 수 있다. 위 예시 테이블을 다시 보자. 가령 “샌프란시스코에 사는 남성 사용자들이 Wikipedia 편집을 한 횟수는?”과 같은 쿼리가 있을 수 있다. 이 쿼리 예시에는 도시(San Francisco)와 성별(Male)이라는 두 가지 차원이 포함된다. 각 차원별로 아래와 같은 바이너리 배열이 생성되는데, 여기서 배열 인덱스 각각은 해당 행이 쿼리 필터 조건에 부합하는지 여부를 나타낸다.
::
San Francisco (City) -> rows [1] -> [1][0][0][0]
Male (Gender) -> rows [1, 2, 3, 4] -> [1][1][1][1]
San Francisco (City) -> rows [1] -> [1][0][0][0]
Male (Gender) -> rows [1, 2, 3, 4] -> [1][1][1][1]

그런 다음 쿼리 필터는 이러한 두 배열에 대해 AND 연산을 실시한다.
::
[1][0][0][0] AND [1][1][1][1] = [1][0][0][0]
[1][0][0][0] AND [1][1][1][1] = [1][0][0][0]

그 결과, 행 1만 스캔 대상이 된다. 이런 식으로 필터링된 행만 검색함으로써 불필요한 부하를 방지하는 것이다. 이러한 바이너리 배열은 압축하기도 매우 쉽다. 이러한 검색 인덱싱은 OR 연산에도 사용할 수 있다. 어떤 쿼리가 San Francisco 또는 Calgary을 필터링하는 경우, 배열 인덱스들은 차원값별로 다음과 같을 것이다.
::
San Francisco (City) -> rows [1] -> [1][0][0][0]
Calgary (City) -> rows [3] -> [0][0][1][0]
San Francisco (City) -> rows [1] -> [1][0][0][0]
Calgary (City) -> rows [3] -> [0][0][1][0]

그런 다음 두 배열에 대해 OR 연산이 수행된다.
::
[1][0][0][0] OR [0][0][1][0] = [1][0][1][0]
[1][0][0][0] OR [0][0][1][0] = [1][0][1][0]

그 결과, 쿼리는 행 1과 3만 스캔한다. 대형 비트맵 셋에 boolean 연산을 실시하는 이러한 접근방식은 검색 엔진에서 널리 사용된다.




쿼리 언어
===================================

Expand All @@ -122,4 +120,4 @@ Druid의 네이티브 쿼리 언어는 JSON over HTTP이며, 주요 쿼리는
* Sum, Min, Max, Avg 등의 집계 연산
* 차원값 검색

하지만 이 외에도 SQL을 비롯한 다양한 언어로 이루어진 쿼리 라이브러리가 생성·공유되고있다.
하지만 이 외에도 SQL을 비롯한 다양한 언어로 이루어진 쿼리 라이브러리가 생성·공유되고있다.
12 changes: 6 additions & 6 deletions docs/_sources/discovery/part01/structure.rst.txt
Expand Up @@ -12,7 +12,7 @@ Metatron Discovery는 metatron 운용 서버에 적재된 데이터 소스나

데이터 프리퍼레이션
=====================================
데이터 프리퍼레이션은 원천 데이터에서부터 데이터를 정제 및 가공하여 메타트론으로 적재하는 기능을 제공합니다. 데이터 프리퍼레이션에 대한 자세한 설명은 :doc:`../part07/data_preparation_overview` 를 참조하세요.
데이터 프리퍼레이션은 원천 데이터에서부터 데이터를 정제 및 가공하여 메타트론으로 적재하는 기능을 제공합니다. 데이터 프리퍼레이션에 대한 자세한 설명은 :doc:`../part07/index` 를 참조하세요.

.. figure:: /_static/img/discovery/part01/prep.png
:align: center
Expand All @@ -26,7 +26,7 @@ Metatron Discovery는 metatron 운용 서버에 적재된 데이터 소스나

데이터 스토리지
======================================
데이터 스토리지는 메타트론 내에서 사용될 메타트론 엔진에 적재된 데이터를 관리합니다. 데이터 관리 기능에 대한 자세한 설명은 :doc:`../part02/data_management_overview` 를 참조하세요.
데이터 스토리지는 메타트론 내에서 사용될 메타트론 엔진에 적재된 데이터를 관리합니다. 데이터 관리 기능에 대한 자세한 설명은 :doc:`../part02/index` 를 참조하세요.


.. figure:: /_static/img/discovery/part01/datasource.png
Expand All @@ -46,7 +46,7 @@ Metatron Discovery는 metatron 운용 서버에 적재된 데이터 소스나

워크스페이스
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
워크스페이스는 조직 내에서 사용할 워크북, 워크벤치, 노트북을 권한에 따라 관리할 수 있습니다. 워크스페이스 기능에 대한 자세한 설명은 :doc:`../part03/workspace_overview` 를 참조하세요.
워크스페이스는 조직 내에서 사용할 워크북, 워크벤치, 노트북을 권한에 따라 관리할 수 있습니다. 워크스페이스 기능에 대한 자세한 설명은 :doc:`../part03/index` 를 참조하세요.


.. figure:: /_static/img/discovery/part01/workspace.png
Expand All @@ -55,7 +55,7 @@ Metatron Discovery는 metatron 운용 서버에 적재된 데이터 소스나

워크북, 대시보드 그리고 차트
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
워크북은 대시보드와 차트를 PPT 처럼 작업하고 한 워크북안에서 여러 대시보드를 보고 프리젠테이션 하는 방식입니다. 워크북 기능에 대한 자세한 설명은 :doc:`../part04/workbook_overview` 를 참조하세요.
워크북은 대시보드와 차트를 PPT 처럼 작업하고 한 워크북안에서 여러 대시보드를 보고 프리젠테이션 하는 방식입니다. 워크북 기능에 대한 자세한 설명은 :doc:`../part04/index` 를 참조하세요.

.. figure:: /_static/img/discovery/part01/dashboard.png
:align: center
Expand All @@ -71,11 +71,11 @@ Metatron Discovery는 metatron 운용 서버에 적재된 데이터 소스나

노트북
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Machine Learning 기반 고급 분석을 수행할 수 있습니다. 노트북 기능에 대한 자세한 설명은 :doc:`../part05/notebook_overview` 를 참조하세요.
Machine Learning 기반 고급 분석을 수행할 수 있습니다. 노트북 기능에 대한 자세한 설명은 :doc:`../part05/index` 를 참조하세요.

워크벤치
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
SQL 기반 분석을 수행할 수 있습니다. 기능에 대한 자세한 설명은 :doc:`../part06/06_01-workbench_overview` 를 참조하세요.
SQL 기반 분석을 수행할 수 있습니다. 기능에 대한 자세한 설명은 :doc:`../part06/index` 를 참조하세요.

.. figure:: /_static/img/discovery/part01/sql.png
:align: center
Expand Down

0 comments on commit ea85dd6

Please sign in to comment.